Microsoft Promises a 'Whale' for GPT-5, Anthropic Delves Inside a Model’s Mind and Altman Stumbles

Microsoft promise ‘whale-size’ compute for a GPT-5-tier model, and say the end is not in sight for scaling the power of AI. Google ship models and a fascinating paper, while Anthropic unveil the inner workings of large language models. Meanwhile Sam Altman is forced to repeatedly apologize, Ilya Sutskever leaves, and GPT-4o is pushed back. My reflections on all of the above, and details you may have missed from each paper.

AI Insiders: https://www.patreon.com/AIExplained

Kevin Scott Talk: https://www.youtube.com/watch?v=2bnayWpTpW8
Mark Chen Hint: https://x.com/GaryMarcus/status/1790122337058119725
Noam Comments: https://twitter.com/polynoamial/status/1676971506969219072
Anthropic Scaling Monosemanticity: https://transformer-circuits.pub/2024/scaling-monosemanticity/index.html?s=09/
https://www.anthropic.com/news/mapping-mind-language-model
Ilya Leaves: https://twitter.com/ilyasut/status/1790517455628198322
Then Jan Leike: https://x.com/janleike/status/1791498183543251017
And Logan Hints: https://x.com/OfficialLoganK/status/1791652970670747909
Altman Apologizes: https://x.com/sama/status/1791936857594581428
https://www.forbes.com/sites/antoniopequenoiv/2024/05/21/sam-altman-apologizes-to-scarlett-johansson-over-openai-chatbot-voice-she-called-eerily-similar-to-hers/
And Her Delayed: https://help.openai.com/en/articles/8400625-voice-chat-faq
Superalignment Starved: https://fortune.com/2024/05/21/openai-superalignment-20-compute-commitment-never-fulfilled-sutskever-leike-altman-brockman-murati/
https://openai.com/index/introducing-superalignment/
Gemini Updated Paper: https://storage.googleapis.com/deepmind-media/gemini/gemini_v1_5_report.pdf
And Prizes: https://x.com/JeffDean/status/1793026704757190697/photo/2
Google AI Studio: https://ai.google.dev/aistudio

Business GenAI Consulting: https://theinsiders.ai

Non-hype Newsletter: https://signaltonoise.beehiiv.com/

AI Insiders: https://www.patreon.com/AIExplained