TheChek News Flow

BARRON'S 2026 TECH ROUNDTABLE — FULL TRANSCRIPT Published: March 27, 2026 Participants: Glen Kacher (Light Street Capital CIO), Brent Thill (Jefferies Senior Software Analyst), Morgan Samet (Lingotto Innovation Managing Partner), Ben Reitzes (Melius Research Head of Technology Research) Interviewer: Alex Eule, Barron's

KACHER: The emergence of AI as an architectural computing change is driving the industry forward. The largest companies with the most capital to spend are saying this is the most important thing happening in our industry. Semiconductor demand is extremely strong, and demand for AI compute cycles is still very strong.

SAMET: The market potential for what AI can do in not just the digital economy but also the physical economy is massively underappreciated. It isn't software that is being disrupted — it's labor. Labor markets are on the order of $10-20 trillion, whether in the digital or the physical economy. Trust and security are incredibly important, but the opportunity is massive.

REITZES: The certainty is that the rate of change will be accelerating. I see more disruption ahead, not less.

THILL: Two things are certain. First, AI will be globally adopted at the board level — not the IT level. Multiple board members I've spoken with have said they're all in, reorienting their companies across every industry. Second, infrastructure: Microsoft, Google, and Amazon are all seeing incredible demand they can't keep up with.

THILL: They are acting offensively. When JPMorgan Chase said it had reduced total operational spend for support but put that money back into revenue-generating roles, that's a board-level topic. If you can cut costs and improve revenue, that is an offensive move.

KACHER: What is amazing about AI is that the industry has built tools that are incredibly productive for the single user. It is very approachable for a CEO. Everyone has had an individual personal experience with this technology. The gap we are all struggling with is that it takes time to build tools for teams. That is where the doubt comes — in applying these tools for large numbers of people to use together.

THILL: In some cases, delays have already happened. CoreWeave had to push out $8 billion of capex in Q4, and had issues getting physical buildings up and live. The demand is exceptional, but schedules are shifting materially. It can come down to a lack of PVC piping that could delay a project by months.

KACHER: Every time we have seen a major architectural change filter through the tech industry, it takes a decade or more. Salesforce was founded in 1999. It took 15 years for roughly 25% of computing cycles to shift to the cloud model, and another eight years to get the next 25%. Today's software isn't going away. You will see AI adopted on top of existing applications, and incumbent software players will build some of those applications.

SAMET: You are seeing this in consumer spaces first because there are no barriers to entry. Executives are operating from a place of both fear and opportunity. People act faster when you pair fear and opportunity.

THILL: Multiple AI vendors are going to win. Intuit uses more than 20. Gemini is better for this, ChatGPT better for that, Anthropic better for this. You will see companies try to find a framework or governor of LLMs. Everyone thought they would become commoditized; now they are becoming more specialized. The embrace of many models inside a system that can orchestrate their use is the preferred route. One AI company coming in and crushing the entire organization is probably not going to happen.

REITZES: This news cycle is like nothing I've ever seen, and I was an analyst in the dot-com bubble. There is a lot of shoot first, ask questions later. Investors need to be aware of both the narratives and the fundamentals.

KACHER: The most conflicting narrative is from tech bears who say AI is so powerful it will destroy all software — but simultaneously say there is no ROI coming, and therefore semiconductor demand will crater. You can't destroy an industry of incredible value if the product doesn't do something incredibly valuable. It is a very bizarre conflict.

REITZES: The tokenization of labor is occurring. AI has the ability to be a PhD in your pocket, a doctor, an analyst, an accountant. It will augment many positions and make people doing these jobs much better. The unfortunate thing is that prior labor disruption cycles took 10-20 years; this one feels like it is taking place over 10-20 minutes. You are already seeing a pause in hiring for knowledge workers.

SAMET: It is going to happen much faster than people think. But there is also far less friction in learning something new. You will see, especially among younger people, more entrepreneurs launching smaller businesses as a result of these tools.

KACHER: Every one of these transitions took 20-plus years. The iPhone came out in 2007; by 2013, smartphones had reached 50% penetration. We always overestimate the speed at which these transitions disrupt existing industries. The Armageddon of employment is probably overstated in terms of speed.

REITZES: Many software companies don't realize how quickly they may need to move to consumption models rather than charging per seat. That is what we call a negative mix shift.

THILL (SNOW): Snowflake is a huge winner. Growing close to 30%, with widening margins and the best margin expansion of any large-cap software company. One primary competitor in Databricks — a two-horse race. You can run every AI technology in Snowflake's platform safely. It is effectively a control tower for all your data, with its own AI agents, partnering with all major cloud vendors.

SAMET (NET / NOW): The most underappreciated company in this area is Cloudflare. It sits in front of approximately 20% of the web between its CDN and security. Most valuable now is its physical edge infrastructure — servers that sit closer to users. As you think about the transition to edge and agentic AI, latency and geography matter. Revenue has accelerated materially in the past four quarters. We also own ServiceNow and Databricks (private).

KACHER (FROG): We like JFrog — a system of record and artifact-management platform used by 80% of Fortune 100 companies and many leading AI labs. We think it will generate more than $2 of FCF per share in 2027. FCF growing 40% a year. Stock likely to trade at 30-35x FCF, implying 50%+ upside. The market has incorrectly identified this business as at risk from AI. In fact, it is an AI winner.

REITZES (IBM): If I had to pick a stock in software, it's IBM. There is a perception that IBM's mainframe business could be threatened by AI — we don't think so. In a world where we're short of compute and CPUs, it may benefit on-premises compute. The majority of IBM's software follows a consumptive and instance-related pricing model, not SaaS — a better place to be. Long term, there is a quantum computing call option. Quantum and classical computing will have their moment in the early 2030s. We are more cautious than Wall Street on Microsoft and Adobe.

KACHER (ASML / MKS): We're big fans of ASML — absolutely necessary for making advanced GPUs, XPUs, and all chips powering the AI buildout. We also like MKS. Advanced semiconductor packaging should continue to grow faster than front-end wafer-fab over the next two years. MKS will have better than $13 of EPS power in 2027 on a modest recovery scenario, trading at 25-27x — implying $350-400 stock, or 60-80% upside. Key supplier to TSMC.

REITZES (INTC): We need more foundries and more advanced chip-making capabilities. Intel has those assets. Lip-Bu Tan is quite an amazing executive — he possesses the skill set to do what Lisa Su did at AMD: fix the product business and create optionality for the foundry. The US government is a shareholder and President Trump is supportive. Intel also has hidden packaging assets doing very well right now.

SAMET: It is astonishing. Nvidia continues to rerate down — it makes no sense given the fundamentals. It has more to do with market structure: 7.5% of the S&P 500 is Nvidia, which pushes limits on how much certain funds can invest.

KACHER: Consensus expects 75% EPS growth in 2027 and 30-35% in 2028. The stock is trading at 17x consensus 2028 estimates. Compare that to Apple, struggling to grow earnings by 10% a year and trading at mid-20s multiple. We own Nvidia and like it a lot. We also own Broadcom for similar reasons — a competitor in the inference space, incredibly well-run.

REITZES: At Nvidia's GTC conference, Jensen Huang gave an order number for $1 trillion in hardware revenue. There is meaningful upside in every quarter through end of 2027.

SAMET: Perhaps the market misunderstands the level of Nvidia's software play. You can't build a robotics company or an autonomous-vehicle company without them.

REITZES: The Mag Seven are under assault from market rotation. The overwhelming issue is the lack of free cash flow for many of them — Alphabet, Amazon, Tesla, Microsoft, and Meta. On the horizon, Anthropic, OpenAI, and SpaceX could come public — potentially the Mag 10 or Mag 11. The question of which stock to sell to make room is hanging over the whole group.

THILL (META / AMZN): Meta will earn more than $30 per share in the next year. Put a 25-30x multiple on that and you have a materially higher stock price. Meta's world is so immersive and different that it will have a seat at everyone's consumer table. The cheapest name right now is Amazon. The death of AWS is greatly exaggerated. Amazon is now at the same EV/EBITDA multiple as the beginning of last year for Alphabet — at or below the average multiple of Walmart. That one seems oversold.

SAMET (CATL): The Chinese battery maker CATL. The cost and quality differential between CATL and the rest of the world is massive — even Tesla uses CATL cells. The market is focused on geopolitical tensions, which are real, but in the short term there is no solution without CATL.

THILL (PCOR): Procore Technologies — vertical software for construction management. Partnered with Nvidia to bring AI to construction. Construction is the most manual, highest-waste industry on the planet. Great mid-cap story vertically focused on global construction.

KACHER (CLBT): Cellebrite — dominant player in digital forensics, providing software to law enforcement worldwide. Going to generate $1 per share of FCF in 2027, growing FCF at 30%. Stock has 50-75% upside. Just introduced an AI product helping law enforcement analyze communications across the entire suite of applications on a phone — solves crimes, increases speed of investigations.

SAMET: Two things. One — a new consumer platform or product that unlocks a major everyday use case beyond current AI assistants. Two — more technology in the physical economy, particularly autonomous vehicles. The technology is solved; it is happening faster than people think.

THILL: Why is capex continuing to go materially higher? That will stunt software growth and continue to benefit infrastructure, semis, and hardware over software. This has been happening every year and the broken record plays on: semis keep outperforming.

REITZES: Data. The big surprise will be how much data all of this requires — more flash, more hard drives, more data storage systems both for cloud platforms and on-premises. AI makes your data more valuable.

KACHER: The debate will be, can we keep spending at this rate? The numbers will get larger for the AI buildout, and the percentage of free cash flow that the Mag Seven are spending will become a much bigger issue. Those doing the spending will get tougher questions asked of them.

END OF TRANSCRIPT Source: Barron's Tech Roundtable, March 27, 2026 — Alex Eule