Dan Ives: Most Disconnected Call I've EVER Seen

Video Description

Book a call with me to grow your business with YouTube - https://calendly.com/ikostiy03/dc Gain lifetime access to my AI course - https://ailcourse.replit.app/?channel=k I'm looking for UK AI Companies to partner with - Email [email protected] I generated 45M views + helped businesses generate $100k's from YouTube in 2024 Follow me on twitter - https://twitter.com/DavidCarbutt_ Contact me here for business or sponsorships opportunities - [email protected] This accountant saved me a fortune. If you're in the UK, book a call with Ben for accounting or bookkeeping. https://calendly.com/donnellanbouchier/20mins Want to join a private stocks group? Click here -https://www.fejeremy.com/2024-app-pltrbtszd 📖 Reccomended Palantir Substacks Arny Trezzi - https://arnytrezzi.substack.com/ Either Square - https://eithersquare.substack.com/ This whole segment is basically the “software is dead” narrative getting stress-tested, and the answer is: some software is absolutely dead… but it’s the feature-companies, not the platform-companies. Dan Ives is arguing the selloff is disconnected because markets are pricing enterprise SaaS like it’s about to get nuked by chatbots, while ignoring the boring truth of enterprise reality: installed base, procurement friction, multi-year contracts, integrations, compliance, and switching costs that turn “disruption” into a slow grind, not a guillotine. The SAP “dark fiber” comparison is the right fear—overbuild without demand—but Dan’s counter is also fair: 1999 was PowerPoints and fake business models; 2026 is Apple/Google/Microsoft/Meta writing the checks with real cash, because usage is already here and the ROI signals are starting to show up. Then Ark adds the nuance that actually matters: platform vs wrapper. If your product is basically “a nicer front-end to GPT/Claude” (or a single AI feature that a model provider can bundle in one update), you’re one release note away from irrelevance—Cursor is the example they’re hinting at: great product, terrifying dependency. But if you’re the platform with distribution + workflows + data + governance (Microsoft, Salesforce, SAP, ServiceNow), you’re Walmart in the e-commerce era: you don’t need to win the model race, you just need to absorb AI into the core product and sell it to the same customers who already live inside your stack. That’s why the “Agentforce + data products at $1.4B ARR growing 100%+” point is so important: it’s proof that “legacy” companies can hide Palantir-style growth inside their base—markets just don’t trust it yet. So the real split forming is simple: wrappers and thin vertical tools get commoditized; thick platforms with distribution get upgraded. And Demis basically tells you what the next battleground is: reliability becomes the moat. As AI goes from “cool demo” to “systems running banks, insurers, hospitals,” the winners won’t just be the most capable—they’ll be the ones who can offer robustness, security, guarantees, auditability. That’s the enterprise endgame: agents everywhere, across surfaces (browser/phone/desktop/glasses), but only the providers who can prove “this won’t hallucinate your compliance into a lawsuit” get deployed at scale. So if you zoom out, this episode isn’t bullish on “all software.” It’s bullish on software that owns workflow + distribution + trust, and bearish on anything that can be copy-pasted into a bigger platform’s roadmap.