“95% Einstein, 5% Mr. Bean”: Jev’s Creator on Why AI Can’t Automate Anything Reliably (Diogo Almeida)

Almeida argues reliable AI automation requires calibrated "decision models" like Jev, not chat-tuned LLMs, which remain structurally unreliable despite benchmark gains.

• Jev's launch metrics: 40M views on launch video, 1M+ users/day, 1 trillion tokens/day, a third of Fortune 500 adoption, and an $870M Series A at $7.5B valuation within three weeks.
• Core thesis: chat-tuned LLMs are "95% Einstein, 5% Mr. Bean" — jagged and uncalibrated — so they can't be trusted to automate reliable decisions inside software.

Themes, sentiment and reasons are AI-written paraphrases, checked so they never quote an episode. They may contain errors. Not investment advice. Episodes remain the property of their creators.