Mato just raised pre-seed Read more
Podcast charts
Published by James Spiro
Join me as I discuss issues relating to Israel, tech, media, and news. Sometimes with a guest, sometimes solo. www.thespirocircle.com
On the charts
Every published chart this podcast appears in, in the snapshot behind this page. Each one links to the chart it came off.
The Mato Topic Intelligence Platform does not report this podcast as charting in a published category in this snapshot.
From the feed
The latest episodes published to this podcast’s own RSS feed. Titles and descriptions are the publisher’s.
Uzi Krieger has jumped from planes, descended into underwater caves, and skied slopes steep enough to unsettle most people’s stomachs. When I asked what draws him to it, he put it down to a philosophy he lives by. “I think it’s a theme in life of being a little bit of an explorer,” he told me. “I don’t want to say ‘living on the edge’, but trying to see how far you can go, how hard you can push, and do it in a very structured way.” For Krieger, now General Manager of Cloud Security and Exposure Management at CrowdStrike, risk is not about recklessness. “It’s not just about jumping off a cliff or diving hundreds of feet underwater,” he explained. “It’s more about taking on a challenge and trying to see how far you can get… That’s how I’ve approached extreme sports, or work, or almost everything I do in life.” The idea of “structured risk” is interesting to me. I learned it personally at a young age when obtaining my SCUBA license: Plan your dive, dive your plan, I was told. It’s something that also grounds my decisions as I grow my independent media venture after departing from corporate newsrooms last year. Krieger was the right person to speak to about this idea and how he applied it across his career. He started in corporate America’s telecom sector, running a business unit inside a large, publicly traded company, until he left to pursue something else. “Very early on I realized that moving fast and being able to get really good results is something that can only happen in Startup World,” he explained, which is also what attracted him to Israel’s startup ecosystem. He went on to found and run several companies, most notably Reposify, an external attack surface management company which CrowdStrike acquired in 2022. CrowdStrike's leadership asked him to help bring more “founder DNA, startup DNA” into the organization. Krieger credited the mentors of his early startup years with teaching him that line between risk and recklessness. “They had a very unique combination of very strong conviction and an extreme, structured, almost scientific way of how you get to being successful,” he told me. It comes down to preparation, not appetite: “You want to take risk, but it’s not just any risk. Where’s the passion? Are you really all in? What is the plan? Are you actually building a plan, or are you just trying to wing it?” The new speed of risk That same instinct now shapes how CrowdStrike approaches its own market, which has hit $243.3 billion, and a stock price circling within distance of an all-time high. Exposure management, which is the practice of identifying and prioritizing an organization’s most exploitable weaknesses, was largely a background chore before AI adoption pushed it into boardrooms. “It’s no longer a question of ‘why do we need this,’” Krieger said of the shift. “It’s ‘how do we move 10 times faster? How can we be ahead of the curve?’” This mindset gets him safely out of a cave, or back on the ground with both feet, and it’s what separates a security strategy from a scramble. But it requires work. “You really need to build that muscle. It’s not natural. People are risk-averse, but how do you transform risk aversion into something that allows you to take risks, enjoy the journey, and do it in a way that you become successful?” His question stayed with me until long after our meeting. The AI era is asking everyone to take risks. Company pivots, career changes, market shakeups… all of them will require us to make decisions or take actions that will impact our personal and professional lives. “It boils down to basic human nature and being a good person. People want to do good; people want to be successful. It doesn’t matter if it’s in a small company or a big company,” he concluded. “I think [it’s about] not losing the edge that you have with a startup, and taking that into a larger organization, and bringing that impact.” [Watch a preview: Why this CrowdStrike manager skydives, cave dives, and calls it “Structured Risk”] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Well, here it is: Episode 100 of The Spiro Circle. When I recorded the first episode 18 months ago, I had a rough idea of what this show could be. I wanted it to be a place where long-form conversations about Israeli tech, geopolitics, or society could get broader attention. Now, with a growing audience across 20 countries and partnerships with international media channels like Forbes, I am proud to see what it is becoming, and that these conversations are getting the attention they deserve. Startup Nation yearned for long-form English conversation, and I am pleased to be helping fill that missing gap. For the 100th, I brought in Nitay Milner, co-founder and CEO of Orion Security, and Emily Fontaine, Global Head of Venture Capital at IBM Ventures, to talk about why every major company should rebuild its data security strategy in this AI era. I find it fitting that our conversation on how a large corporate company saw potential in a scrappy Israeli startup was positioned as the 100th episode. Thanks to all who have joined to share their stories so far, and thank you to all those who keep tuning in. Onwards to 200. - JS The data loss prevention industry has organized itself around a relatively stable threat model. A disgruntled employee may copy files to a USB drive. Or a phishing attack could expose credentials to an outside attacker. Maybe someone accidentally CCs the wrong person on an email with sensitive information. Solutions focused on perimeter controls or policy rules were imperfect, but understandable when the threats were often caused by humans. But that model is no longer sufficient. According to research by Cyberhaven Labs, whose 2026 AI Adoption and Risk Report analyzed data movements across 222 companies, nearly 40% of all AI interactions now involve sensitive data, with the average employee inputting proprietary information into an AI tool once every three days. Threats don’t disagree insofar as they disaggregate, accelerate, and, in some new cases, remove the human from the equation entirely. Nitay Milner, co-founder and CEO of Orion Security, offered a framework for understanding how the landscape has changed. In a recent episode with Emily Fontaine, Vice President and Global Head of Venture Capital at IBM, which backed ORION’s $32 million Series A in February 2026, he described the three categories of traditional data leakage that the DLP industry was built to address: human error, malicious insider activity, and external attackers. “[Perhaps] I accidentally did ‘Emily@IBC’ and not ‘IBM’,” he said, by way of illustration, “and sent the entire board deck to the wrong person. That happens a lot. I call it keeping honest people honest.” The second category, he noted, involves deliberate exfiltration, what he called the “Snowden” scenario. The third involved external actors penetrating an organization and quietly siphoning data over time. Milner co-founded Orion Security in 2024 with CTO Jonathan Kreiner. It aims to replace traditional DLP (data loss prevention) tools with an automated, context-driven platform. Using LLMs and specialized AI agents, the platform continuously detects and analyzes data loss indicators in real time, capturing context for content sensitivity, data lineage, user identity, behavioral intent, and environmental purpose. New anatomies, new leaks Traditional categories of DLP remain relevant, of course. But Milner identified two new vectors that are reshaping the problem. The first is data extraction into third-party AI: employees uploading sensitive documents like earnings calls, customer records, or source code to unmanaged AI platforms before those materials are cleared for external use. “Taking the earnings call, which is super sensitive data, before the earnings call report was published, and uploading it to like an unmanaged ChatGPT or Claude,” he said. “It’s very, very sensitive, and data is now in the hands of a third party that you don’t have any agreement with.” The second vector is data exfiltration not by humans at all, but by AI agents operating inside the enterprise. “AI agents doing human work inside the organization, having access to super sensitive data,” Milner explained. “Think about an AI agent email assistant that has access to your Google Drive, takes the entire customer list and sends it to the wrong person. That’s data exfiltration by AI.” The scale of this emerging risk is becoming measurable. According to a 2026 Cloud Security Alliance report, 82% of organizations already have AI agents operating in production environments, while only 17% enforce runtime access controls consistently across those deployments. A separate finding from Proofpoint’s 2025 Data Security Landscape report found that 32% of organizations identify unsupervised data access by AI agents as a critical threat. Fontaine framed the underlying problem as one of movement, not just volume. “Data is a huge asset that must be protected, more so than ever before,” she added. “It’s moving across clouds, it’s moving across applications, agents, ecosystems. It’s moving across so much more than it was ever before. And we have to make sure it’s secure, that it is governed correctly.” IBM Ventures, the strategic investment arm of IBM led by Fontaine, operates a $500 million fund focused on AI and quantum technologies — and its bet on Milner and the team reflects IBM’s belief that the arrival of large language models has fundamentally broken the assumptions on which traditional data security was built, creating entirely new leakage paths that legacy DLP tools cannot detect. A new era for DLP That governance challenge is exactly what today’s DLP industry is no longer built to handle. Policy-based systems depend on known patterns, like a credit card number matching a regex or a file name triggering a rule. They cannot, by design, interpret context or whether a particular data movement constitutes a legitimate business action or an exfiltration event. Milner’s breakdown offers enterprises a useful starting point to identify which of the five categories (legacy or AI-era) represents the highest unaddressed exposure in a given environment, and build from there. “Understand your organization,” he concluded. “Based on this real data, get to decisions, train your employees, and help them understand how to use data safely.” [5-minute preview: Is your AI leaking company secrets? IBM & Orion Security explain] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
For two decades, the winning formula in consumer tech was simple: maximize time on screen. Last month that formula got a multibillion-dollar price tag attached to it when Meta agreed to pay roughly $17 billion to settle a multistate lawsuit alleging Facebook and Instagram were deliberately engineered to be addictive to teenagers. Now, one Israeli investor thinks the backlash could create an opening for a very different kind of consumer product, and for a new generation of Israeli founders. Oren Charnoff is the Co-founder and a General Partner at Sticker Ventures, a Tel Aviv early-stage fund that invests exclusively in Israeli B2C startups. He sees the settlement as validation of a thesis his fund was already investing behind, based on changing consumer habits among young people. “We want to find more companies whose goal is not to increase time spent on the app,” he said. “We hope to see more ‘IRL’ things being built by Israelis.” He points to two U.S. examples he admires. The first, Tin Can, is a landline-style phone that plugs into the wall, paired with an app that lets parents approve who their kids can call; Charnoff says it’s doing roughly $80 million in annualized revenue. The second is Board, a digital game console built entirely around in-person play. “The only way that you can interact with it is with another person with you,” he said. Closer to Sticker’s own market, Charnoff cites Edikted , the Israeli-founded Gen Z fashion label, which The Wall Street Journal recently reported as achieving roughly $460 million in revenue: “They crush it on retail. They crush it for offline discovery.” Startup Nation’s advantage in the changing market The shift makes it perfect timing for Israel’s tech DNA to tap into adapting markets. Startup Nation has accumulated decades of expertise in adtech, gaming, fintech and cybersecurity, producing a generation of founders trained to measure, test and optimize. Add to that a new type of internet-native, young immigrant to Israel who is a product of globalization, and those skills can be unleashed on the world’s biggest GDP category: consumers. “The same quants who do anomaly detection in cyber can optimize the budget, channel and yield of consumer growth,” Charnoff said. He traces the lineage directly to Israel’s earlier tech waves. “Adtech is one of the founding mothers and fathers of B2C,” he said, pointing to the exodus of former ironSource employees now building consumer companies of their own. The craving to steer away from online apps and toward “In Real Life” extends into dating apps, both in his portfolio’s orbit and out of it. Companies are moving away from “unlimited swipes whose goal is to maximize dwell time” toward an AI concierge model that delivers one or two curated introductions a week, rather than an endless feed of profiles. According to Charnoff, 38% of Sticker Ventures’ deployed capital sits in the Health & Wellness category - framing the appeal generationally: millennials were “an experiment of unlimited internet access”, he told me, while Gen Z is driving “a huge trend to get back into IRL... experiences.” Charnoff and Sticker Ventures aren’t betting that the attention-economy giants disappear. But the “hipster move” into B2C investments is placing new bets on the next generation of consumer winners built for the time people spend away from their news feeds, not the time they spend scrolling through it. “B2C’s been a big part of Israel’s ecosystem for a long time,” he concluded, listing success stories like Waze, Oddity, Superplay, eToro, Wix, Simply, MyHeritage, and others. “We’ve always been doing this… I just think with AI, there’s a renewed interest to do it… It’s just way more attractive to be a B2C entrepreneur now.” [Preview: Why building B2C in Israel is suddenly a "hipster move"] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Guy Katsovich and I sat across from each other in our shared studio a few weeks ago, discussing what happens when podcasting moves from the media ecosystem into the campaign trail. Let’s set the scene: We are both former journalists, now running separate podcast operations (one in English, one in Hebrew), and talking about what happens when an interview format becomes part of an election strategy. Katsovich is the co-founder and partner at Fusion VC, a pre-seed platform and a longtime Israeli podcaster, and expects that trend to accelerate as election day approaches. “If you look at the trend, I would assume... they’ll be interviewed by more and more podcasters... the closer we get to election day,” Katsovich told me. The 2024 U.S. election showed just how powerful podcasts could become as a political campaign tool, with candidates increasingly opting for long-form conversations that offered more time and unfiltered depth - something traditional television can rarely provide. Israel’s next election, scheduled for October 27, is already generating a growing political podcast ecosystem, from dedicated election shows to established interview programs. Prime Minister Netanyahu’s campaign has leaned into a circuit of small, friendly podcast studios in recent weeks, trading hard questions for a more intimate, folksy format that lets a candidate control the narrative in ways a more adversarial Channel 12 or Channel 13 panel never would. Katsovich sees the same pattern and traces it to an obvious source: “Everybody now follows Trump’s steps and tries to interview [on] as many podcasts as they can.” His show, The Guy Katsovich Podcast , now reaches, he told me, “more than 200,000 people... every month,” almost entirely inside Israel. He’s hosted Yair Golan, Gadi Eisenkot, Bezalel Smotrich, Simcha Rothman, and with an election bearing down, he expects the guest list to grow: “I might do a roadshow for all the politicians that wanna come on the podcast.” Israel appears to be the latest country that is redistributing its political access during election cycles. Katsovich’s read is that as October nears, “those podcasts will carry more and more weight in how people make their decisions,” because a long-form sit-down is “a real opportunity to really meet and see a politician for one hour” in a way a ninety-second clip cannot replicate. But the format only works if it stays open. “All the politicians that came to my podcast didn’t ask for preparation, and there were no preconditions,” he told me, before admitting some guests want questions in advance, or want to see the cuts before publication. “This is like... kills the whole point of a podcast,” he said - and I agreed. “If you wanna really do an open conversation, this is not the format” for anyone trying to manage their message. Startup Nation’s election role But access is only half of the equation. What politicians choose to talk about matters too. When I pushed him on whether topics like tech and AI sovereignty will actually move votes — a theme this campaign has already flirted with, and one I have discussed on another podcast with ILTV — he identified a key thesis for election messaging. “If in America you vote for economy, in Israel you vote for security, and that’s it.” [Watch me on ILTV: Maayan Hoffman and I explore whether Startup Nation should weigh in on Israel's first election since October 7] Whereas the American mantra once promised that “it’s the economy, stupid”, Israel sees things a little differently. As Israelis go to the ballot box for the first national election since October 7, high-tech, cyber, and defense are all what he called “intercollided” into a single binary: are you the guy who keeps the country safe, or not? That’s the paradox sitting underneath this whole moment. Hebrew podcasting is gaining exactly the kind of reach and intimacy that could reshape how Israelis size up a candidate — and it’s happening at the same time as candidates are working out how to use that reach without actually being tested by it. [Watch a preview: Israel’s podcast boom could change the next election] The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
America’s fight over data centers is becoming a mainstream political staple. Seven in ten Americans say they don’t want a data center near their home, and more people now believe these facilities harm their community than help it, according to recent polling cited by industry trackers. Opposition has more than doubled across 49 states over the past year, and in the first quarter of 2026 alone, an estimated $130 billion in data center projects were blocked or delayed by local resistance. It’s not even a culture war. The backlash against them cuts across party lines: Republicans tend to object on tax incentives and grid strain, whereas Democrats oppose on environmental and resource use. The reasons may differ, but their mutual opposition is a rare instance of bipartisan alignment in American politics. Into this landscape steps Phinergy. It’s an Israeli cleantech company whose aluminum-air backup generators have become an alternative to diesel in the AI infrastructure race, seeing early support from players in an impressive Google- and Microsoft-led innovation consortium. I spoke to its CEO Emmanuel Levy, who laid out a thesis that the “NIMBY” reflex against data centers collapses once three specific, solvable grievances are addressed. “People are caring about [them] and are afraid... whether it is pollution, whether it is noise, or it is water consumption,” Levy told me. “If you’re able to explain why a new data center can come without this price on the shared resources, you’re able to bring the debate back to where it should be.” It’s a clean, testable claim. In Colorado Springs earlier this year , a proposed data center drew objections almost identical to the ones Levy’s technology is designed to neutralize: noise and air pollution from backup diesel generators, and water consumption. The developer countered with a water-efficient, closed-loop cooling design and pledged 60 to 100 permanent jobs. But it didn’t defuse the fight. Longtime residents cited broken promises tied to a chip plant and a cryptocurrency mine previously sited on the same land, and the developer told local reporters he had never encountered this level of public opposition, despite having built a similar facility elsewhere without incident. The complaints were the same three Levy names, yet the outcome was not. Why do they need to be in our neighborhoods, anyway? Those opposed to data centers don’t want them “in their backyard” - but that’s precisely where they need to be. Whereas training AI models can in fact be anywhere, the inference stage of AI - that’s getting the answers back to the customer who inputs prompts - needs strong latency to ensure smooth communication. And those are fast becoming critical infrastructure, which means backup generators need to be precisely alongside those initial centers. “More and more services are AI-based; they need their AI to be as close as possible to the market,” Levy explained. “So that’s the reason why inference in this case is localized next to communities.” Think of it like being closer to your internet router at home: a reasonable comparison, until you realize that McKinsey forecast confirmed that by 2030 there will be an additional 140 gigawatts of power consumption by data centers. It means that the equivalent of 140 nuclear power plants will need to be set up in four years in order to manage the computing power. “You understand why there is a bottleneck,” he added. Another question arises: whose job is it to close this trust deficit? Levy describes a cooperative, three-layer division of labor: new technology needs to be “socialized” to the market, regulators need to update rules that still legally define backup power as diesel-only, and developers need to carry the message to communities and win consent. Phinergy was selected from a field of more than 70 proposals by the Net Zero Innovation Hub, a consortium formed by Google and Microsoft specifically to source alternatives to diesel backup power, with the pitch vetted privately by senior industry advisors (including a former chairman of Vertiv's technology board and a former Microsoft data center technology executive) before hyperscalers would take it seriously. "It's rarely cold calls," Levy said of how deep tech actually reaches Google and Microsoft. "Without that, it's very, very difficult for people to take you seriously." Phinergy’s pitch for cleaner data centers is the start of the battle. Removing diesel noise and emissions from their footprint retires one legitimate grievance from a list that has grown to include broken trust, property values, and a broader unease about ceding land and resources to AI infrastructure. International polling on the same backlash lists the objections in similar order: water, energy, land use, noise, air pollution. But it still needs to overcome the lingering challenge of a lack of community consent. Clean power may be necessary to end the data center fight. On the evidence so far, it isn’t sufficient. You can learn more about data centers and how Phinergy can fix the bottleneck in the entire episode. [For now, here is a preview: The NIMBY problem with AI data centers, explained] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
I’ll admit it upfront: I’ve tried alternative meat on my own barbecue, and they’ve never been exactly a slam dunk for me. So when I sat down with the Chunk Foods founder and CEO Amos Golan for this latest episode, I wasn’t chasing a hype piece on the wonders of foodtech. I’ve written those stories before. What I wanted to know was whether the economics of fake meat had finally changed. Turns out there is a number that can make that argument: $6.83 , the price of a pound of ground beef in the US as of July 2026. Beef steaks averaged $12.80 a pound back in May , up 16% year-on-year; the national cattle herd is sitting at its smallest size in 75 years. These Bureau of Labor Statistics numbers show a worrying trend for families and young people seeking affordable protein sources. Chunk offers clean-label, plant-based whole cuts that look (and cook) just like beef. And today’s price rises may be a way for the fake meat industry to undergo somewhat of a comeback. Golan has been watching this trend as closely as I have. “You don’t need to just talk about the need for the price of [fake] meat to go down,” I said to him. “You can also talk about the price of meat going up.” His answer: “Exactly.” He’d already walked me through the math before I said it. “Ground beef in the US used to cost $4 [per pound]. Now it’s $5, $5.36. A steak is $10 to $20 a pound.” That was true when we recorded this episode in June, but beef has kept climbing since, which is, for those of us doing the weekly shop, the whole point of his pitch. Chunk’s bet was never to compete with beef on ideology. People like me who genuinely like eating meat aren't suddenly going to stop because someone tells us it's better for the planet. “The experience needs to be good enough. The price needs to be good enough, ideally cheaper than meat,” he added. He was blunt that this is a demographic shift, not a moral one. “People are less altruistic, I’d say, at least externally, in how they talk about the food they consume.” Ultimately, beef isn’t getting cheaper anytime soon. Forecasts put 2026 beef price increases as high as 18%, and analysts aren’t projecting relief into 2027. Chunk doesn’t need to win me over on taste alone. It needs the gap between $6.83 and its own shelf price to keep doing the argument for it. It might be time that the alternative meat market comes back for seconds. And those at the dinner table may finally be ready for what’s on the menu. [PREVIEW: Rising beef prices are doing Chunk Foods' marketing for them] The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Glilot Capital’s internal AI agent has a name and a full-time job - Elliot. It reads deal memos, surfaces founder meeting summaries, coordinates diligence advisors, and routes work across one of Israel’s most prominent VC funds. Amit Spitzer, Glilot’s CTO and CISO, built it largely on his own, without a dedicated budget, while running the fund’s security operations in parallel. “We need some super agents, some super tools, some ‘super employee’,” he told me. “That were able to connect to all the tools and all the processes and basically help streamline every manual work that used to be done up until that point.” Arik Kleinstein, Glilot’s co-founder and managing partner, added that before the agent existed, an investment decision required five to ten diligence calls conducted manually. Now, the firm can screen advisors, make contact, present a company, and schedule calls at a pace that would have been unimaginable three years ago. And since Glilot is a seed and early-stage investor, operating in a market where the window between a promising founder and a closed round can be very short. An ‘employee’ like Elliot can help VC funds find their next gold mine. Yet for all of its assistance, Kleinstein maintains that the agent cannot do the actual job. “If you invest very early, you invest basically in two things,” he said. “One, the team, and in their imagination. And AI is not a team, and AI doesn’t really have a good imagination.” The difference between what AI accelerates and what it cannot replace carries broader implications for how the industry should think about the technology it is rushing to adopt. Even I, as a writer, admit to using the technology in my work; transcriptions take seconds instead of hours, and I have tweaked my agents to be harsh in their review of my work before I submit to a (human) editor. It means I have more time for research and deeper conversation with my guests. In Glilot’s instance, the fund is as committed to AI as any in the market, and it evaluates every investment, in the words of managing partner Lior Litwak, “through a GenAI lens.” But Kleinstein and Spitzer make a sharp distinction between AI as infrastructure and AI as judgment . While AI in our workflows is indeed transformative, the ability to assess an unproven founder's leadership potential remains out of reach — at least for the investing Glilot does. “All the analytical part, to look at the technology, to look at the markets, to look at the products, all the checklist that every VC is doing… that can be significantly augmented with AI,” Kleinstein said. “But to assess the quality, the leadership, the ability of a founder, many times a first-time founder, never done it before, to actually build a team, that’s something where AI just cannot help you.” Spitzer adds another dimension to this argument. Describing his CISO position as “part of my soul, not just my experience,” he stated that the people who get the most out of Elliot will not necessarily be the most technical members of the team. They will be the ones who know what to ask, and how. “You still need the experience of the person,” he said. “Because now you need to know what to ask the AI, how to ask it, how to navigate in the path of AI. And that skill is important with AI and before AI, but now it’s making it even more critical.” That idea of “tribal knowledge” inside an organization or team to carry the qualities needed to thrive in the AI era was discussed in a previous episode of The Spiro Circle . AI may have access to an organization's information, but it doesn't automatically understand which pieces of that information matter, or spot what an experienced human employee would notice that isn't written down. This reframes the debate about AI and employment and the fear that it will “take our jobs”. Currently, it appears that there will be a big shift in which part of a job it takes, with a recent IIA/Zvriran study on Startup Nation suggesting that while AI isn’t the primary reason companies are reducing their headcounts, the technology is cited by 10% of companies as the cause of hiring pauses - a threefold increase from six months ago. In Glilot’s case, Elliot absorbs the data-gathering, the scheduling, the pattern-matching across thousands of founder meetings. What it leaves behind is soft skills, such as the ability to sit across a table from someone who has never built a company before, and decide whether to bet on them. “The emotional part, the charisma, that’s especially important if you’re a very early-stage investor,” Kleinstein concluded. “That’s much more important than any data I can crunch.” [Watch a 5-minute preview: AI can do the diligence. But can it pick the founder?] The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
The week Gal Ordo joined Amazon Web Services as the Product Leader for AWS Security Hub, he did something his colleagues thought was unnecessary. Instead of using the internal templates that AWS provided to get the product up and running, he insisted on installing it from scratch the way a customer would - no shortcuts or help from the team that built it. But then he said it took him four to five days. “It was so hard for me to install,” said the co-founder and CPO of Native Security. “And I am within AWS! I’m the product leader for that. What does it mean for a customer going in the first time trying to use that? At best, they give up; at worst, they misconfigure it, and then they don’t have active security coverage.” That initial moment planted the first seed. But the second came a month later, when Ordo attended his first Executive Briefing Center session on behalf of Amazon. There, a room of executives from Fortune 100 and Fortune 50 companies had one message for the young product leader sitting across from them: “Why are you showing us the problem after it happens? Why can’t we just prevent it?” “I remember them telling me, ‘Why do you even allow me to get to that place? Why don’t you just give me the knob to say this can never happen in the first place?’” Ordo recalled. It was those two experiences, operating a product too complex for its own builders to operate, and a customer too exhausted to keep reacting, that became the founding logic of Native Security. The Tel Aviv-based cloud security startup emerged from stealth in March 2026 with an official $42 million in funding (he says they have already secured more, up to $55 million), and a collection of Fortune 100 customers already running its platform in production. According to industry research, 80% of cloud breaches stem from misconfigurations — resources left exposed, permissions set too wide, settings that drifted after an engineer’s late-night change. Among companies that use multiple cloud providers, only 33% have a unified security strategy across them. Ordo’s argument to me is that the tools to close that gap have always existed inside AWS, Azure, Google Cloud, and Oracle - but enterprises cannot operationalize them at scale. Native’s answer is what Ordo calls a control plane, a platform that lets security leaders specify an outcome in plain language, such as: “My sensitive data should never be exposed to the internet.” It then translates it into enforceable technical controls across all four major cloud providers simultaneously, using those providers’ own native capabilities rather than adding another external layer. Before any policy goes live, the platform simulates its impact by surfacing which services might break, or which cross-cloud dependencies might be disrupted, so security teams can act without fear of taking down production. “The building blocks are there, they’re strong, they’re powerful,” Ordo said. “But they’re hard for customers to use.” Ordo co-founded the company with Amit Megiddo, CEO, who led Amazon GuardDuty, and Eyal Faingold, CTO, who served as VP of Cloud Security Products at Check Point. Between them, the three founders have shipped security products used by tens of thousands of enterprise customers globally. The initial $31 million Series A was led by Ballistic Ventures, whose roster includes Phil Venables, former CISO of Google Cloud. He joined the round and, later, the board, even calling Native’s approach “the next big evolution in cloud security.” The company currently employs around 50 people across Tel Aviv and the United States, drawing talent from IDF cyber units, Palo Alto Networks, Wiz, and CyberArk, with a target of 90 by the end of 2026. The founding question was whether the knowledge they accumulated inside those organizations could be turned into something different. Notably, knowledge of where the tools broke down, where customers gave up, and where the cycle of detect-and-react became self-perpetuating. “There is a reason this hasn’t been done before,” he concluded. “It’s incredibly difficult to achieve. The technical problems we’re solving are very hard.” [5-MINS PREVIEW: The founding story of Native Security] The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Most VCs will tell you they invest in people . Push them further, and you’ll hear the same formula of exceptional team, large market, and right timing. It is recited so often it has become meaningless. But Yuval Ariav, Founder and Managing Partner of Symbol, doesn’t give me that answer. Ask him what he looks for in a founder, and he’ll acknowledge that integrity and grit matter, but then he sets that all aside. “If I had to put one overarching theme for us, it’s depth,” he told me. “We like to see that founders went to the validation threshold the capital market requires — and then way beyond it. Because what that tells us is that they are really interested in the problem, that they really understand it.” Much of the theme of today’s conversation is about depth, and how to avoid “non-deep” thinking (I try to avoid the term “shallow”). For Ariav, the ability to go deep is the willingness to go so far into a problem that you exit the comfortable zone of apparent mastery and enter the uncomfortable zone of knowing how much you don’t know. “We like to see people whose level of expertise forces them to say, on some things, ‘we don’t know.’ Because that’s how you really know they know their stuff.” Symbol is a $50 million pre-seed and seed fund Ariav co-manages with Racheli Kogan, and bets on founders working outside the mainstream of Israeli venture, in sectors the local capital community has traditionally dismissed. We spoke soon after Business Insider named Symbol to its 2026 Seed 100, the first time an Israeli fund has appeared on a list that includes Sam Altman, Accel, and Greylock. The ranking’s stated criteria, which highlighted those who identify technologies before they become mainstream categories, are basically Symbol’s entire thesis. The problem is that depth is vanishing There’s a paradox at the center of Ariav’s argument: That the quality he bets on is becoming scarcer precisely as building companies gets easier. Because of AI, he notes, “building products has never been easier, faster, cheaper.” When there are more founders, lower barriers to entry, and more buzz, the very environment that Symbol is making a contrarian bet on is becoming full of less intellectual effort. “The capacity for depth… it’s not just that you understand something deeply,” he said. “It’s that you are fine with sitting on your ass at home and spending six hours diving into god-awful McKinsey super boring reports to understand some fundamental truth. That is a skill set that is rapidly fading from the world.” Research found that average sustained attention on a digital screen dropped from 150 seconds in 2004 to 47 seconds by 2024 . Ariav, who teaches a course at Columbia University on data, AI, and society, watches this in real time. “Our collective attention span is going way down,” he said. “At a time when the world is more complex than ever.” Contrarianism isn’t the answer Here is where Ariav diverges from the standard VC contrarian pitch. Being reflexively contrary, he argues, is just as lazy as following the consensus. “Being reflexively contrarian is also easy. You look at Twitter [X], and you just add ‘no’ in the beginning of whatever somebody says.” What Symbol looks for is different: founders who have gone deep enough to discover something the market hasn’t seen. “When you meet the best founders, at some point they will say something to the effect of: we’ve observed something the market doesn’t understand. At that point, they shine a spotlight on a hidden truth.” For Symbol, it is about finding the pre-consensus thinking across a world that is no longer operating in the very depths that depend on new discovery. And with a world that changes as AI becomes more embedded in our lives, the ability to think, explore, and innovate is becoming a commodity in and of itself. [Watch a preview: The "Depth" Problem: Why great tech founders are getting rarer] The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
In cybersecurity, hesitation can be the most expensive decision a company makes. Shahaf Galili knows this not as a management theory, but as a lived operational reality forged across 22 years running offensive cyber units inside the Israeli military. “You always need to make decisions when you don’t have all the information,” says Galili, co-founder and CEO of Mars Security, a Tel Aviv-based threat detection startup that raised $9 million in 2025. “The most dangerous thing to do — if you’re investigating a breach, if you’re conducting an offensive operation, if you’re trying to decide which product line to invest in — is to wait. Targets are moving and evolving. There’s a lot of chaos and multiple unknowns.” It is a philosophy shaped by his real-world experience in the Israel Defense Forces, confronting actual life-or-death consequences. Galili left the IDF at 39, relatively later than most who leave after their mandatory three-year commitment, which he considers an advantage. The extra years gave him something that fast-track civilian careers rarely produce: a visceral comfort with operating under conditions that would paralyze most executives. “The more chaos that you live through,” he said, “the better you will thrive in chaos.” That thesis now underlies how Galili runs his founding team and how he thinks organizations should respond to a security breach in progress. His core conviction is that an imperfect action consistently outperforms a perfect decision arrived at too late - or one not performed at all. “I believe that an action, even if it’s the wrong one, is always better than not acting,” he explained. “An action will help you understand reality. You need to act… and then have the flexibility to understand that you made a mistake and change.” I pointed out that there is something almost Shakespearean in how he thinks about cybersecurity. Hamlet, history’s most famous illustration of the cost of indecision, understood perfectly what needed to be done but died precisely because he couldn’t bring himself to do it. The play is a 400-year-old case study in strategic paralysis, and one that Galili would recognize immediately as a security failure as much as a human one. History, it turns out, doesn’t repeat itself - but we all know it can rhyme. Delay was Hamlet’s ‘fatal flaw’. In cybersecurity, it remains one of the most common that can still cost enterprises their data. This is not merely philosophical for Galili. The detection gap that Mars Security is built to close between the moment an attacker enters an organization, and the moment the organization realizes it is partly a technology problem and partly a mindset one. Security teams have been conditioned to gather more data before acting. Attackers, meanwhile, are moving across identities, cloud environments, SaaS platforms, and corporate networks in hours, not days. “You need to be good all the time,” Galili said. “The attacker has one advantage: he has the initiative. If you take this from him and start being proactive, looking all the time at how they’re evolving, they don’t have a chance.” The best piece of advice he ever received from a commanding officer? He doesn’t hesitate: “Love the chaos.” The ability to embrace chaos in 2026 is as applicable to building a startup as it was to running operations in the years before - even if contexts and technologies evolve. And the leaders who thrive are never the ones who waited for more certainty before moving. They’re the ones who always kept moving. The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
While I was recording an episode with Toni Samson, co-founder and CEO of Arte Security, she told me it was important for companies and CISOs to “find your Mona Lisas.” It made me remember how a few thousand kilometers away, the actual home of the actual Mona Lisa had learned that exact lesson the hard way. Last October, thieves walked out of the Louvre with crown jewels worth more than $100 million. French investigators later found that only one of the two cameras covering the break-in point was even working, and staff didn’t have enough screens to watch the footage that did exist. The whole operation lasted minutes. Imagine that: The Louvre, home to the most famous painting on earth, couldn’t tell you in real time what was happening to the jewels twenty feet away. It didn’t know where its own “Mona Lisa” was, so to speak. And a few days ago its gallery finally reopened - this time, without the stolen jewels. That’s the trap Toni was warning me about, except she wasn’t talking about paintings. Her metaphor was directed at CISOs who need to know the most important assets in their enterprise, and make sure those are more protected than just applying blanket protection over everything. “Find your Mona Lisas,” she said. “This is the most important. You will not be able to close everything. It’s too much. It will never be fast enough.” The value of each asset varies by company. A shoe manufacturer’s factory floor matters more than its HR files. A bank’s Mona Lisa is the data itself, sitting in very specific places. “The question now is: How can I protect my most critical things? [Because] not everything is the Mona Lisa,” she said. “Not everything needs to be protected the same.” This is where Arte’s own work gets specific. The company helps enterprises identify what parts of their data require more attention (and protection) than others. It then specializes and tailor-makes a solution that helps them protect their Mona Lisas from theft or hacking. The company was founded with co-founder and CTO Asaf Ohayon at the end of 2025, and Toni herself comes from a background in the Israeli Ministry of Defense as Director of Critical Infrastructure & Data Center Cybersecurity. Another example may be a hyperscale AI data center, where its Mona Lisa isn’t necessarily a database. It might be the chiller controller. Bad actors wouldn’t even need to breach a firewall to take an AI cluster offline. All they would need to do is make the room too hot to run. This sounds obvious until you actually try to do it. Security researchers estimate that roughly a third of large businesses can see less than three-quarters of their own assets at any given time: Think of it as the digital equivalent of a museum that isn’t sure how many rooms it has, let alone what’s in them. I asked her to help me quantify it: how do you actually know which door is the one with the painting behind it? To distinguish between the Mona Lisas and what I called “The James Spiro Original Scribble”. Her answer was to identify what you’d protect first if you could only protect one thing, and build outward from there. “Make sure your Mona Lisa is protected,” she said, “and put it as number one priority.” Every laptop or every forgotten API endpoint is indeed an attack surface or entry point. And for a long time, the instinct in cybersecurity has always been to try to lock all of them at once. But Toni’s point is that this instinct is now outdated, because when attackers can use AI to move at machine speed, treating every asset as equally precious means treating none of them as precious enough. The Louvre is now spending close to a billion euros to build the actual Mona Lisa her own dedicated room. Most businesses won’t get that budget. CISOs will have to actually know where their Mona Lisa hangs before someone else finds out. [Watch a preview: “Not everything is the Mona Lisa” — Cybersecurity priorities, explained] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Israel’s cyber numbers from 2025 were staggering. We all know the big ones: Google acquired Wiz for $32 billion. Palo Alto Networks spent $25 billion to absorb CyberArk. ServiceNow bought Armis for $7.8 billion. When the final tally was complete, strategic mergers and acquisitions in the cybersecurity sector had reached $81 billion - more than four times the volume of the prior year, according to data compiled by Notable Capital and Morgan Stanley. For Oren Yunger, Managing Partner at Notable Capital and one of the architects of the firm’s annual Rising in Cyber report, the significance of those deals runs deeper than just their valuations. “The ceiling has kind of been shattered,” he told me. “You can build businesses that are just massive. You can do things that we, in the past, thought were unimaginable in cybersecurity,” he added. “Companies should get acquired between $200-400 million dollars — that’s the best you can do. Now we’re seeing companies that really are paving the way to the next set of companies to go and build bigger and stronger.” For much of the last decade, cybersecurity exits were measured in hundreds of millions, not the tens of billions we see today. The Wiz deal alone, which became the largest high-tech exit and business deal in Israel's history, poses a new question to a new generation of founders: If they could become a $32 billion business, what is possible for us ? The M&A wave of 2025 coincided with a shift in how enterprises think about the AI era and the risks that come with it. In 2024 and early 2025, the defining question for enterprises deploying AI was whether they could build agents capable of autonomous action. Yunger now argues that that question has largely been answered and a new one has emerged. “The big question that we’re seeing is: can you put it in production, can you scale that agent, can you trust its operations?” he said. “Security is just a huge part of this question that needs to be answered at enterprise scale.” The Rising in Cyber 2026 report, which surveyed nearly 150 chief information security officers from the world’s largest companies, puts precise numbers on the gap between deployment and protection. Seventy-one percent of respondents said their organisations already have AI agents running in production environments. But only 11% described their tooling to secure those agents as mature. That gap is where Yunger sees the next wave of investment flowing, and where the next generation of large cybersecurity companies will be built. The broader cybersecurity sector is projected to reach $255 billion by 2029, up from $153 billion in 2025, according to IDC estimates. Early-stage investment is accelerating: Series B rounds grew 74% to $3.3 billion in 2025, with average deal sizes jumping 75% to $49 million. In a year when most software categories saw venture funding decline, cybersecurity’s earliest rounds were the only segment to grow year over year. The momentum has continued into 2026. CrowdStrike acquired SGNL, Palo Alto Networks purchased Koi, and Sophos acquired Arco Cyber — all in the first half of the year. The platforms that spent 2025 making transformational acquisitions are now targeting the AI-native capabilities they need to stay competitive as the threat landscape continues to evolve. Microsoft, which dominates five of the largest cybersecurity verticals, is adding capabilities faster than at any point in its history, with CrowdStrike and Palo Alto doing the same. But their acceleration has not crowded out newer startups. If anything, it has raised the stakes for founders who can identify the problems that the large platforms have not yet solved. “Security companies today are answering those questions,” Yunger said, “and the ones to follow will eventually accompany every single technology shift that is happening in the market and will continue to do so for as much as I can think of.” Yunger claims the thesis is confirmed by Satya Nadella, Microsoft’s chief executive, who he said has described engineering as converging into four enduring disciplines. One of them is security engineering: a category he considers permanent regardless of how AI reshapes the rest of the technology industry. “Doesn’t matter how AI is affecting our markets, what jobs AI potentially will threaten and maybe change… Security is here to stay.” For founders building in the space today, that is both a reassurance and a challenge. The ceiling “has been shattered” and the market is expanding faster than we all predicted. Large companies are acquiring quickly, and the venture dollars are flowing earlier. What founders may be asking themselves now is if they can also build something the platforms may have no choice but to pay to absorb. And Wiz has shown them that the answer may be worth $32 billion. [Watch a preview: The Cybersecurity Ceiling Just Got Shattered] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Israel isn't fighting one information war. It's fighting three. At least, that’s the takeaway I had when I read Digital Warrior : Inside Israel’s Battle for the Narrative After October 7 by Rachel Lester this week. Rachel served in the International Branch of the IDF Spokesperson's Unit, first on active duty, then in reserves from October 8th, 2023, editing and shaping video content for accounts with over 10 million followers across platforms. You probably saw her videos from the official IDF X account, or from her personal Instagram, @Rachel.In.Reserves Thinking back to the early days following the October 7 massacre, what struck me most was the impossible communications challenge the IDF faced. Her job, and the job of the army, was to explain multiple things to the world at once. “I think that Israel faces social media challenges that no other army and no other country in the world face,” she told me. “I think that we have the unique challenge of trying to reassure our civilians that we are safe and that the army is in control, while at the same time conveying to the world these are our enemies, this is what our enemies are doing right now… this is what our enemies just did: committed the largest attack on Israel in our history. And if we don’t fight them, then they’re gonna do it again and again, as they’ve said.” One country, three completely different messages One of the central arguments in Digital Warrior is that the IDF isn't running one narrative. It’s running three, simultaneously, to three audiences - and the messages don’t just differ in tone, they actively contradict each other. * To the local, Hebrew-speaking audience - the message is reassurance: we are strong, we’re handling it, you don’t need to worry. * To the Arabic- and Farsi-speaking world - the message is closer to a warning: we are strong, we’re watching you, and your own leaders are the ones corrupting you. * And to the international, English-speaking audience , the posture flips entirely: we are at risk, we are under attack — because that’s the only framing that earns Israel the right to defend itself in the eyes of the world. Project strength to one audience and vulnerability to another, at the same time, and you’d think it’d collapse under its own weight. Rachel told me it basically does, sometimes. The three departments work almost entirely independently of each other - there’s no room for someone to coordinate the tone across all three. “Sometimes it works, and sometimes it doesn’t,” she said. The moment she paused I considered the paradox Israel found itself in, and thought about why it didn’t apply to Hamas’ messaging strategy. So I put it to her directly: Didn’t Hamas actually pull off the version of this that Israel is still struggling with? A fear campaign aimed at the region, running alongside a sympathy campaign aimed at the West without the contradiction ever really catching up to them? “That’s an interesting point I hadn’t considered before,” she said. Western outlets, she pointed out, never really broadcast the moments when Hamas spokespeople promised “to do October 7th again and again.” That footage exists, as Rachel saw firsthand, but it just doesn’t travel the way photos of dead children (or P.O.D.C., a term she coined for the book) do. As the war dragged on, Hamas’ messaging was contained inside its Arabic-language threats, but its English-language grief campaign spread across the world. Somehow, the two rarely collided in the same feed, in front of the same audience, at the same time. This was not true for Israel, whose messaging sometimes came out slowly or inconsistently - and whose critics were quick to highlight these contradictions. This is more than a media-strategy story It would be easy to file this under another “PR problem” Israel has to overcome, and move on. But I don’t think that’s what it is. Our conversation kept circling back to how, in a war where legitimacy determines whether you’re even allowed to keep fighting, the coherence of your story and its narrative matters as much as the truth of it. That challenge is only becoming harder now that AI makes it easier to manipulate and dismiss authentic content. During our conversation, we discussed the “liar's dividend”, a term coined by legal scholars Bobby Chesney and Danielle Citron for exactly this phenomenon: real evidence getting waved away simply because fake evidence is now possible. She mentioned an instance in 2023 when journalists doubted the validity of footage released by the IDF. And earlier this year, genuine footage of Netanyahu was mistaken online for an AI fabrication. So the tech doesn't even need to be actively used against Israel. Its existence alone is often enough to doubt or undermine its digital efforts. That’s the tension I’d encourage you to sit with if you pick up Digital Warrior : not “is Israel’s PR bad,” which is the question everyone already has an opinion on, but “can any democracy actually hold three contradictory messages together in an age when information is everywhere?” Digital Warrior is out now . You can follow Rachel on Instagram here . For transparency: I have no financial relationship with this book or its sale. This recommendation is unpaid. [Preview: We discuss “How the IDF Talks to Israelis, Enemies, and the World — All at Once”] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
On July 8, the head of the National Highway Traffic Safety Administration sent a letter to the autonomous vehicle industry saying that AV developers had shown a “clear pattern” of driverless cars blocking ambulances and fire trucks, ignoring flares and flashing lights, and in some cases driving directly into active emergency scenes. While not technically naming Waymo, it was undeniably a focal player in the robotaxi reckoning - saying there was “a functional insufficiency” among automated vehicle developers regarding “a pattern of interference with first responders”. At the same time, a different conversation has been unfolding inside the AI industry. Some of its leading figures have begun looking beyond engineering for answers. Anthropic recently launched its Faith & AI Covenant Roundtable to discuss how best to infuse morality and ethics into AI, while OpenAI's Head of Strategic Futures, Dean Ball, sparked debate after saying he had begun studying the Talmud to better understand AI policy. Dr. Tal Cohen, co-founder of Drive TLV and managing partner of Next Gear Ventures, joined me to discuss these issues. In an interview recorded just after the NHTSA letter, Cohen used Waymo as the clearest example of what he calls “The Habitat”: the missing institutional layer of trust, permissioning, and accountability that must exist around an AI system before it’s allowed to act with consequential impact in the world. “Two, three years ago, I was irrelevant, because there was no capability to talk like that,” Cohen said. But AI ability is expanding, and it is clearly starting to outpace the governance structure meant to contain it. “The capability is expanding,” he added, pointing to Waymo’s rapid deployment. “Suddenly, you have a gap between what the capability can provide and what we are lagging as a society.” Cohen’s main argument is that the industry has spent its energy on the wrong bottleneck. Public debate about AI has focused overwhelmingly on physical infrastructure constraints related to chips, energy, and data centers, or on the capabilities of the models themselves. But he told me the actual constraint is a broader and less visible vacuum around the governance that authorizes an autonomous system to act, who reviews what it did, and who has the standing to revoke its authority when it gets something wrong. “So then the question is: who’s gonna own the habitat that’s gonna authorize Waymo to go into crime scenes or not?” He argues society will hand over billions of consequential driving decisions before governments build the institutional framework capable of supervising them. In his framing, regulators, municipal transportation departments, and NHTSA itself are unlikely to build that infrastructure fast enough on their own. He expects, and believes the moment demands, some hybrid of public and private coordination to define a constitution of sorts for autonomous systems. “We’re gonna live in the centuries or decades of habitat construction,” he said. “People really don’t get it yet.” A manuscript he's been circulating, The Case for Habitat , argues this point exactly: That without a system or code in place, organizations face an uncomfortable choice. “You either put somebody in the basement, don’t let it do what it can do… or let it destroy your business,” he concluded. In Cohen's words, the capability is “shiny,” but the Habitat is “boring.” Yet history suggests that the boring is often what determines which transformative technologies succeed. [5-Mins Preview: Can Judaism solve AI alignment?] Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Every day, billions of dollars move through the global payments system as people buy things online. And every day, a portion of those transactions gets reversed. Not through fraud in the traditional sense, but through disputes: a cardholder tells their bank they don’t recognize a charge, or that a service wasn’t as described, and the money often comes back to them almost immediately. Multiplied across the entire e-commerce economy, that adds up to an estimated $125 billion a year . It’s a leak most consumers never think about, and most of fintech rarely discusses. In the AI era, however, as users continue to delegate AI agents to conduct financial transactions online, it is becoming the latest hurdle global payment companies need to overcome. Ofir Tahor, co-founder and CEO of Justt, has spent the past six years building a company around that gap. His explanation of the problem starts with its age. “This is a mechanism created about 60, 70 years ago,” Tahor said, describing how card networks originally designed chargebacks to protect cardholders in a world of physical stores and mail-order catalogs. A customer could go straight to their bank rather than the merchant, flag a transaction as unrecognized, and get reimbursed while the burden of proof shifted entirely onto the business to explain why it should keep the money. Tahor explained that perhaps the most noteworthy aspect of the industry is how almost nothing else in e-commerce still works this way. “If you take companies like Shopify, which did a revolution in the e-commerce world, and Stripe, which did a revolution in the payment processor world, chargebacks stayed behind,” he said. Checkout, fulfillment, customer support, and fraud screening have all moved to real-time, largely automated systems. But dispute resolution remains a paperwork exercise: a merchant compiles documentation, sends it to their payment processor, which forwards it to the cardholder’s issuing bank, like Chase, Citibank, or Wells Fargo, where a human being reviews it and makes a call. “It’s still very manual-operated,” Tahor said. “It’s somehow stayed behind in comparison to many other processes within e-commerce.” The consequence is a phenomenon known as ‘friendly fraud’: cases where a legitimate transaction gets disputed anyway, whether through genuine confusion or deliberate manipulation of a system stacked in the cardholder’s favor. It’s now the second most common type of fraud globally, and as the ability for agents to buy things themselves only speeds up, Tahor doesn’t expect it to slow down. “It’s easy to report a chargeback, and it’s becoming easier,” he said. The data backs this up. Mastercard’s 2025 State of Chargebacks report, based on research from Datos Insights, forecasts global chargeback volume growing 24% from 2025 to 2028, reaching 324 million transactions annually. It’s a trajectory that was already straining a decades-old system before AI-driven commerce entered the picture at all. The argument that clamping down on friendly fraud will be a net positive for the ecosystem and, in turn, the consumer, is the origin story behind Justt’s name. “It’s from the word ‘justice’, in order to create balance in the ecosystem,” Tahor said. Even though it may sound like the company is out to protect Big Business, it is an attempt to build a system that helps merchants keep money they’re owed while still returning money to cardholders when they’re right. In that world, everyone wins because consumers aren’t left paying those costs. Justt was founded in 2020 and has since raised $100 million. It works with more than 250 global enterprise merchants and over 80,000 small businesses, and was named to Forbes’ 2026 Fintech 50 list this spring, the first chargeback-focused company to make the list. It shows that chargeback management, long treated as a back-office cost center, is being recognized as core financial infrastructure in its own right. Learn about Ofir Tahor, chargebacks, fraud, and building Justt in the preview here: The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Days before Tal Shoham announced that his AI monetization startup Velocity had closed a $27 million Seed round, an independent tracker put a number on something the industry had only been speculating about: ChatGPT ads were showing up in roughly 26.5% of all replies globally, and 49% of replies in the United States . For Shoham, that timing is the proof of concept for his newest company and a sign of the latest change in the AI era. “ChatGPT, of course, has added ads now, which is amazing for us and for the industry, because it’s like a north star that everybody looks at,” Shoham told me. He co-founded Velocity alongside Amir Shaked and Nimrod Zuta, all three of whom are former senior executives at ironSource and Unity. The company is building what it calls a growth infrastructure layer for AI-native applications: an ad network, a mediation and auction system, and a “conversation intelligence” layer that turns chatbot dialogue into structured, privacy-safe intent signals. Basically, it’s helping bring adverts to your favourite AI agent. The round was led by NFX and Red Dot Capital Partners, with participation from Stardom Ventures, Corner Ventures, and Transcend, alongside a roster of gaming and ad-tech angels, including former ironSource co-founder Omer Kaplan. The pitch is a straight transplant of the problem his team spent a decade solving in mobile gaming, with one crucial difference. “Ninety-five percent of the users in gaming will never pay a dime,” he said. “You really want to try to find a way of monetizing those users.” In gaming, a free user costs almost nothing. But in AI, that math is inverted: “Every free user that you have on your AI platform is actually costing you a lot of money on inference, tokens, GPUs, and so on.” That inversion is the reason that AI companies have defaulted to hard limits (two or three free prompts a day) rather than the generous free tiers that built mobile gaming and social media into mass-market platforms. He bets that an advertising layer can fund broader free access without those companies bleeding cash, and that doing so improves retention rather than damaging it. “We have more than 12 design partners live already,” he said. “This doesn’t harm retention, it doesn’t harm engagement, it doesn’t harm conversion to monetization.” But the timing that makes Velocity’s raise look prescient also drops it into the middle of an unresolved trust problem - one that OpenAI itself has been actively renegotiating in real time. ChatGPT’s original ad policy excluded placement near politics, health, and mental health topics, with a standing ban on dating, alcohol, drugs, and gambling. But a June 2026 update already suggested that current advertising categories “may expand over time” to include medical, legal, and financial advice contexts eventually. In other words, the rules of the road are being written after the road has already opened to traffic. It’s a pattern that anyone who lived through Europe’s post-hoc arrival at GDPR will recognize as headache-inducing. I pushed Shoham directly on where that leaves the user. Chat conversations are not basic search queries: they’re often confessional, emotional, and far more revealing than anything a keyword ever captured. “There’s a lot more emotion behind what people are giving these algorithms,” I said. “It’s not just tapping into data points... It’s tapping into a real human feeling.” Shoham’s answer leaned on the compliance muscle memory his team built at ironSource, navigating GDPR and a patchwork of state and platform-level privacy rules for years. “We don’t take any of the private information from the user,” he said. “If you type in something on health, something sensitive, your social security number, or whatever it is, we’re not saving that, we’re not taking that, and we’re not integrating that into the model when we’re trying to find the right ad to show you. We have an abstraction layer that actually abstracts all the sensitive information.” So, whereas search reads your keywords, social media reads your behavior, AI just reads you. Velocity’s bet is that the same compliance discipline that got ironSource through GDPR can keep that power in check… but with ChatGPT's ad rollout already outrunning its own written rules, that's a promise the whole industry is now testing in public. Preview: The Next Google Ads? Inside Velocity’s $27M Bet on AI “Intent” Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Manny Marotta has a theory about why America’s 250th birthday felt subdued. It isn’t purely politics, though politics is tangled up in it. It’s the structure. There are simply too many feeds now, and not enough shared ones. And the ones that do break through to the masses get read as political, whether they mean to be or not. I sat down this weekend with Manny for the second time. He’s the creator and curator of the Live History Project, which takes a couple of accounts on X and posts in real time what’s happening in that moment in history. There’s: 25 years ago - @25YearsAgoLiv e 50 years ago - @50YearsAgoLive 100 years ago - @100YearsAgoLive and 250 years ago - @250YearsAgoLive Right now, that means we’re living through 2001 , 1976 , 1926 , and 1776 simultaneously. He pointed me back to America’s bicentennial in 1976, which he says was one of the only major stories of that year, competing for attention with little more than an Olympic Games. This year, Independence Day landed alongside a World Cup on American soil, an ongoing Iran conflict, a White House renovation project, and an MMA match. It also took place with a media landscape noticeably divided along party lines and contrasting levels of patriotism between political ideologies. “Now we have so many different news cycles, so many different news sources that people are following,” he told me. “It’s just oversaturation.” His 250-years-ago account picked up roughly 300,000 followers and 20 million views in the days around the holiday, almost entirely because a political audience decided it mattered. That’s where the story gets complicated. Manny insists the account isn’t doing anything ideological - he just posts digitized letters and meeting minutes from the Library of Congress that are available to everyone, without commentary. And yet that neutrality is precisely what got it adopted as, he describes, a patriotic rallying point by people ‘on the right’. “A neutral or positive view of not just the American Revolution but American history in general has become, in recent years, sort of right-wing coded,” he explained. “So if you are even reporting in an academic sense what happened, a lot of people do tend to see that as right-wing.” Meanwhile, news outlets covering the holiday split along familiar lines: CNN described the mood as shaping up to be “a big blah.” The New York Times ran an op-ed blaming the Trump administration for deflating the day, then was forced to revise its own headline. The Washington Post called it “an unfortunate metaphor on national divisions.” Disney, by contrast, ran wall-to-wall patriotic programming, and outlets like The Free Press leaned into celebratory content. But Manny didn’t spare the current administration either, telling me the patriotic messaging he’d seen recently during a trip to Washington, DC, centered more on a single political figure than the anniversary itself: “The only America 250 content that I saw were giant banners with Donald Trump’s face on them... nothing about the anniversary itself, more about the person who happens to be president.” His hope, he said, is “to create maybe a simulation of the monoculture that we had in the past,” which is academically sourced, uncaptioned, and a return to the apolitical. So he is trying to hold a neutral center by republishing old letters, in a country where an audience conditioned by fragmentation has decided that the center no longer exists. But a Jefferson draft, posted without a caption, still lands as a statement to somebody. This is my second conversation with Manny Marotta. Watch the first, from February, about the Live History Project’s 2001 account, here . Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Jira tickets used to sit open for years. A medium-severity vulnerability, flagged by a routine scan, could be assigned to an engineer who had bigger fires to fight. It had largely been that way for years: patch the criticals, manage the highs, let the mediums age. When it came to CVEs (Common Vulnerabilities and Exposures), a publicly available list of known cybersecurity flaws in software and hardware, nobody was going to weaponize one rated 5.4. But in today’s world, that’s no longer true. “In 2020, [if] there was a CVE reported and a security hole, it would take more than a year until there was a public exploit,” said Shimon Tolts, CEO and co-founder of Tel Aviv-based cloud security startup Copperhelm. “Nowadays, with Claude and OpenAI and other players, the time has shrunk from one year to one day. So now you treat every CVE, every security issue that you have, as immediately exploitable.” The data confirms what Tolts describes. The mean time between a vulnerability being discovered and its exploitation has dropped from nearly a year in 2021 to just over a day in 2026, with industry projections suggesting the window will shrink to one hour by 2027. Rapid7’s 2026 Global Threat Landscape Report found that what once unfolded over weeks now materializes in days (and in some cases, minutes), with the median time between vulnerability publication and inclusion on CISA’s Known Exploited Vulnerabilities catalog falling from 8.5 days to five. The implications invalidate an entire category of enterprise risk management. For decades, security teams built their workflows around severity scores. The National Vulnerability Database , operated by the National Institute of Standards and Technology (NIST), classified every disclosed flaw as ‘critical’, ‘high’, ‘medium', or ‘low’ - and organizations built their response hierarchies accordingly. Fix the criticals immediately, schedule the highs, and then defer the rest. That model is now under institutional strain: CVE submissions surged 263% between 2020 and 2025, and starting April 15, 2026, NIST announced it would only prioritize enrichment for a narrow subset of vulnerabilities, such as those already on CISA’s exploited list, those affecting federal systems, or those covered by Executive Order 14028. This would leave the majority of newly disclosed flaws without severity scores. “You’ll no longer be able to use the old risk management methodology of saying ‘I’m only going to fix criticals’,” Tolts explained. “Because you’re not going to have a severity anymore.” The shift has a compounding effect. AI models are not only accelerating exploitation timelines, but they are also discovering vulnerabilities at a rate that human analysts cannot process. NIST enriched nearly 42,000 CVEs in 2025 , 45% more than any prior year, and forecasts from the Forum of Incident Response and Security Teams projected a record 50,000 additional CVEs to be reported in 2026 (these figures do not yet account for the accelerating contribution of AI-powered vulnerability discovery tools like Claude Mythos and GPT-5.4-Cyber). Every day, the cyber world is facing more vulnerabilities, faster exploitation, and fewer severity scores to guide triage. But security teams are still largely operating through manual workflows designed for a different era. Copperhelm’s answer is autonomous investigation and remediation, already backed by a $7 million seed round led by TLV Partners and deployed in Fortune 500 environments. The platform uses a proprietary “Context Lake” to structure cloud data across environments, enabling AI agents to continuously monitor infrastructure, investigate threats, and execute real-time remediation without manual handoffs. Tolts describes the practical effect in terms his customers already understand: one client arrived with 10 million open vulnerabilities and two home-made severity categories above “critical” — labels they had invented themselves because the official scale had run out of runway. “Your window of response has shrunk, and you need to autonomously take care of it,” Tolts said. “It’s no longer the case where you can just open a Jira ticket and wait for some engineer to fix it in one year or one month, because now you’re gonna get exploited very, very fast.” Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
In the early days of Teramount, investors kept asking CEO Hesham Taha the same question. He had a platform that could connect chip to chip using light instead of electrons, considered a technical feat he and co-founder Avi Israel had spent years developing. But the problem, he recalled, was everything else. “We thought it was a great idea, see how easy we can connect the light to the chip. Everyone will use that,” Taha said. “But it turns out to be exactly the opposite in the first few years after our inception.” The problem for Taha and Israel was that for nearly a decade, Teramount was trying to solve a problem the semiconductor industry didn't know it would have. The market didn’t yet exist, nor did the supply chain. “The most critical point and the big barrier at the beginning of this journey was, ‘What is the product? What is the use case?’ This is what every investor kept asking us, and we failed to give a good answer.” Years passed, and that question, once unanswerable, just got answered. In April, Molex announced the acquisition of Teramount for approximately $430 million, roughly 7-8 times the $58 million that the Jerusalem-based startup had raised across its lifetime. The exit is a case study in what might be called ‘The Patience Trade’: bet on a technology before the world knows it needs it, endure years of uncertainty, and trust that the market eventually catches up. In Teramount’s case, it took two pivots, one global AI infrastructure boom, and a seed investor willing to see something others couldn’t: Lior Handelsman. Today, Handelsman is a Managing Partner at Grove Ventures, and before that, was a co-founder of SolarEdge - so he himself is no stranger to building technologies into markets that don’t yet exist. When he first encountered Teramount, his instinct was to pass. “There was no market even when I met them at the beginning of 2021,” he said. “And I was pretty much willing to tell them, ‘Look, guys, very nice, but I can’t see the market’.” Handelsman ended up reaching out to senior contacts at NVIDIA, Broadcom, Cisco, and Intel — companies that would eventually need exactly what Teramount was building. “When I told them, they said, ‘That’s a big problem. Connecting fiber to chip? That’s a big problem. We are all going to need that in four to five years’.” Grove led the seed round in 2021, and the next few years compressed faster than anyone predicted. The 2022 generative AI explosion turbocharged demand for the kind of optical connectivity Teramount had spent years perfecting. Co-packaged optics — the integration of optical engines directly with compute chips to reduce power consumption and latency — moved from a niche conference topic to an urgent industry priority. And so Teramount, having spent years building the ecosystem relationships and supply chain partnerships that most competitors hadn’t started, was suddenly indispensable. Taha points to two moments that changed Teramount’s trajectory. The first was 2017, when co-packaged optics began to emerge as a defined technology category. The second was 2024, when AI infrastructure demand made optical connectivity not just desirable but necessary. “This was the major and significant pivot in our journey,” he said. Strategic investors followed the technical validation. AMD, Samsung, and Hitachi all joined Teramount as the company’s direction became increasingly legible to the industry. Handelsman describes the combination of financial investors alongside strategic ones as the signal that a company has crossed a critical threshold: “That’s like a sweet spot. A financial investor is leading the round, saying that there is still upside, and strategic investors, who can all be customers.” For Taha, the Molex acquisition was less a finish line than a pragmatic decision about speed. “We had a great technology, we have a great product, but we need to move fast to match the market speed,” he concluded. Molex, a proven interconnect manufacturer with global production capabilities, offered the industrial scale that the Jerusalem-based startup could not self-assemble quickly enough. The patience trade paid off. The lesson it offers is about endurance, and about finding investors willing to hold the same long view as the founders they back. The Spiro Circle is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Everyone is talking about a water crisis brought on by AI use. But think of this: the data centers powering the AI economy are not built from code. They require concrete, steel, cranes, and the coordination of hundreds of subcontractors across millions of square feet of new floor space. And right now, the construction industry is struggling to keep up. Capital expenditure from the 14 largest publicly owned data center operators globally is projected to approach $750 billion in 2026, up from under $450 billion the year prior. The Stargate Project alone, a multi-year, $500 billion plan to deliver up to 10 gigawatts of AI-ready power, was formed as a joint venture between OpenAI, SoftBank, Oracle, and MGXis and spans multiple U.S. states. Meanwhile, the median cost of building a data center hit $445 per square foot this year, up 7.4% from 2025, with average costs skewed far higher by hyperscale projects. The pipeline is enormous, and yet the pressure on builders to deliver is greater still. Erez Dror has seen this shift from both sides. A structural engineer and former construction superintendent who spent over a decade on job sites in Israel, he co-founded workforce intelligence platform, Genda, which last year was acquired by Buildots after a $5.5 million Seed round. He recently stepped into the new role of VP of General Contractors & Genda at Buildots, which to date has raised $166 million. After the acquisition, the joint entity is now positioning itself as the operational backbone for exactly the kind of complex, fast-moving builds that the AI infrastructure boom demands. “A product executive who worked on the biggest project Genda was on, a $600 million project, took them four years to build,” Dror told me. “He moved to build one of the biggest data centers in the U.S., which was $6 billion — 10x the scale — and they built it in three years. A year less, and 10x times the scale.” The compression reflects a new standard being set by hyperscalers who come from a software-first culture and expect physical construction to behave accordingly. “A person who works for Google and is used to building software that doesn’t break expects to get a building that doesn’t break at the same quality,” he said. “They’re setting a new standard, which I believe will eventually trickle down to everything.” The challenge is that construction remains one of the most fragmented, data-poor industries in the global economy. Unlike a tech organization, where a single executive decision can transform operations overnight, construction is built around individual projects with its own lead, subcontractors, or even its own tolerance for disruption. Change management, Dror argues, is “just a different beast.” That fragmentation is precisely what Buildots is trying to solve. The platform ingests two data streams: weekly 360-degree camera footage from job sites and the project’s 3D building model, to use computer vision to identify what has been built versus what was planned. Genda, meanwhile, tracks where workers are on-site in real time, anonymously, using an app-based system that Dror designed around behavioral incentives rather than hardware. Together, Buildots says the platforms offer visibility into both the work being completed and the labour required to complete it. “We know the output, we know the input… we know what was built, and we know what efforts or how many resources were needed to get there,” Dror explained. “We’re the only solution in the world that can provide you with the full picture. Not even at scale — just to provide that.” In April, Buildots formally launched a new product category, which it is calling “construction intelligence” . It frames itself as the operational platform for an industry that can no longer afford to rely on gut instinct and fragmented spreadsheets. And as data center construction starts reached $9.8 billion per month through April 2026 (300% more than levels seen a year ago), the timing for a platform that can turn chaotic job sites into predictable delivery machines has never been better. “When you need to build a facility like a data center that is very detail-oriented, and you need to build it very fast, and every day of delay is millions, if not tens of millions, if not billions, in liquidated damages, you really need to make sure you finish on time and you know what the hell is going on in your project,” he added. Watch a 5-minute preview of this conversation here: Get full access to The Spiro Circle at www.thespirocircle.com/subscribe
Ranking source
Apple Podcasts rankings via the Mato Topic Intelligence Platform.
Observed September 17, 2026. Cached outside the daily freshness window; the positions keep the date they were taken on.
Apple and Apple Podcasts are trademarks of Apple Inc., registered in the U.S. and other countries.
Pairs with
Bring this source into Mato to read its transferable patterns, then turn them into an original show for your own audience.