🧨 Sam Altman Flags Rough Vibes at OpenAI

OpenAI just hit a moment of honesty: a leaked memo shows Sam Altman telling staff the vibes are rough as growth slows and Google gains ground. With revenue concerns and fresh technical setbacks, the pressure is rising. Let’s dive in!

In today’s insights:

LATEST DEVELOPMENTS

Evolving AI: Altman warns staff of rough vibes and slowing growth as Google pulls ahead.

A leaked memo shows Sam Altman acknowledging rough vibes inside OpenAI, with concerns about slowing revenue, technical setbacks, and Google’s strong pre-training results. The note signals a shift from confidence to urgency as staff face the prospect of stalled enterprise demand, hiring freezes, and a renewed push on long-term bets like superintelligence.

OpenAI feeling the pressure is a signal that the easy-growth phase for AI is fading fast: MIT now pegs around 95% of enterprise projects as delivering no real P&L impact, even as hyperscalers pour close to $400 billion a year into data centers. Even OpenAI, which still beats revenue forecasts, runs heavy losses and massive infrastructure spend. For builders and buyers, the message is clear: tools that ship with clean workflow fit, measurable time savings, and fast payback will win budgets, while everything else risks getting stuck in pilot purgatory.

Tired of explaining the same thing over and over again to your colleagues?

It’s time to delegate that work to AI. Guidde is a GPT-powered tool that helps you explain the most complex tasks in seconds with AI-generated documentation.

1️⃣Share or embed your guide anywhere
2️⃣Turn boring documentation into stunning visual guides
3️⃣Save valuable time by creating video documentation 11x faster

Simply click capture on the browser extension and the app will automatically generate step-by-step video guides complete with visuals, voiceover and call to action.

The best part? The extension is 100% free

Check out Guidde

Evolving AI: Meta bets on energy deals to keep its AI plans on track.

Meta is preparing to trade electricity so it can lock in long-term energy deals for its expanding AI infrastructure. The company says new data centers need far more generation capacity than utilities can build without firm commitments. Trading lets Meta support new plants, manage swings in its own usage and sell surplus back into the market. It’s part of a broader shift where large tech firms act as both huge consumers and active market participants.

Meta stepping into energy trading signals that AI is starting to rewrite how grids get financed and built. Data center electricity demand is expected to soar, with AI driven workloads alone forecast to more than quadruple over the next decade, which strains existing generation and transmission plans. Tech giants acting as long term buyers and occasional resellers of surplus can pull new capacity and low carbon projects forward in time, but they also gain outsized influence over where and how new plants get built. For anyone building AI products, this shapes whether the chips you rely on stay online, stay affordable, and stay clean enough to keep regulators and customers off your back.

Evolving AI: Google sets its sights on a giant leap in compute.

Google plans to expand its AI capacity fast, doubling serving capability every six months. Internal teams are pushing for more efficient models, new chips, and tighter integration to meet demand. Leadership points to constraints already limiting growth, with Veo as an example. At the same time, employees question long-term financial pressure as capex surges past operating income.

This is Google trying to turn compute into a long-term moat, not just a bigger bill. A 1000x jump means more people can hit “generate” on stuff like Veo-level video, multi-agent workflows, or fat context windows without waiting in queues or blowing the budget, while smaller teams rent supercomputer-style capacity instead of buying it. The flip side is brutal capex and real bubble risk, which makes this a high-stakes bet on AI demand staying hot for the rest of the decade.

Deepgram surveyed 400 senior leaders on voice AI to map adoption, budgets, and use cases. Compare your voice AI roadmap to $100M+ enterprises and learn where to invest next - human-like agents for customer service, task automation, and order capture - plus benchmarks to guide your 2026 plan.

See the Findings

QUICK HITS

🍌 Google’s Nano Banana Pro generates excellent conspiracy fuel.

📉 Gemini 3 Pro and GPT-5 still fail at complex physics tasks designed for real scientific research.

👏 AI pioneer calls for UK to 'be brave' in tech race.

🏁 OpenAI is ending API access to fan-favorite GPT-4o model in February 2026.

🏛️ The White House has paused a federal order that would have overridden state-level AI regulations.

*partner link