Chamath Drops a Massive Warning скачать в хорошем качестве

Chamath Drops a Massive Warning

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 dcsocialmediadc@gmail.com I generated 45M views + helped businesses generate $100k's from YouTube in 2024 Follow me on twitter -   / davidcarbutt_   Contact me here for business or sponsorships opportunities - dcsocialmediadc@gmail.com This accountant saved me a fortune. If you're in the UK, book a call with Ben for accounting or bookkeeping. https://calendly.com/donnellanbouchie... Want to join a private stocks group? Click here -https://www.fejeremy.com/2024-app-plt... 📖 Reccomended Palantir Substacks Arny Trezzi - https://arnytrezzi.substack.com/ Either Square - https://eithersquare.substack.com/ This pair of clips is basically the AI stack story in stereo: Dan Ives is pointing out that once Nvidia ignites a true “AI factory” buildout, the real compounding happens in the second and third derivatives—cybersecurity, memory, energy, and the enterprise plumbing that makes AI usable and safe—because every new agent, workload, and internal-data connector multiplies the attack surface and forces spend to follow (CrowdStrike/PANW/Zscaler aren’t “nice-to-haves,” they become table-stakes for running AI at scale). Chamath’s twist is even more disruptive: if enterprises realize that “public endpoint AI” effectively creates unavoidable data exhaust—prompts, traces, metadata, sensitive docs—then the pendulum can swing back toward private compute/on-prem or tightly controlled private clouds, not because the cloud failed, but because confidentiality becomes the binding constraint in an agentic world. That doesn’t kill hyperscalers, but it changes who wins the next wave: the beneficiaries aren’t just AWS/Azure/GCP, it’s also the private infrastructure ecosystem (enterprise GPUs, servers, networking, and the software that governs access, auditability, and deployment). In other words, AI doesn’t just increase demand for compute—it increases demand for controlled compute, and the companies that can operationalize AI securely inside regulated, messy enterprises are the ones that quietly end up owning the “application era” when everyone else is still arguing about model leaderboards.