Nothing Much Happens in AI, Then Everything Does All At Once

When it rains, it pours. OpenAI Operator tested and reviewed, with full paper analysis. Perplexity Assistant is useful. Then Stargate, is it all smoke and mirrors? Strong rumours of an o3+ model from Anthropic. Then a full breakdown of Deepseek R1, and what it’s training method says about the state of AI. It’s not open source BTW. Plus Humanity’s Last Exam, and Hassabis Accelerates his AGI timeline.

https://app.grayswan.ai/arena/chat/harmful-ai-assistant
https://app.grayswan.ai/arena

AI Insiders ($9!): https://www.patreon.com/AIExplained

Chapters:
00:00 - Introduction
00:54 - OpenAI Operator
04:53 - Perplexity Assistant
05:15 - Stargate
07:51 - Better than o3?
08:25 - DeepSeek R1 Analysis
12:12 - Training Secrets
15:19 - No More Process Rewarding ?
19:01 - Hassabis Timeline Accelerates
21:22 - Humanity’s Last Exam

https://openai.com/index/computer-using-agent/
System Prompt: https://github.com/wunderwuzzi23/scratch/blob/master/system_prompts/operator_system_prompt-2025-01-23.txt

OpenAI Operator: https://operator.chatgpt.com/
System Card: https://cdn.openai.com/operator_system_card.pdf

There is No Plan: https://x.com/jeffclune/status/1882120726339318007

Perplexity Assistant: https://x.com/perplexity_ai/status/1882466239123255686

Stargate: https://openai.com/index/announcing-the-stargate-project/
https://x.com/sama/status/1882505650594611588
Labour goes to 0: https://moores.samaltman.com/
Noiam Brown Manhattan Project: https://x.com/polynoamial/status/1881833454213767600
Larry Ellison AI Surveillance: https://x.com/TheChiefNerd/status/1882042989184430332
Amodei 1984: https://www.bloomberg.com/news/articles/2025-01-22/anthropic-ceo-says-openai-s-stargate-venture-seems-chaotic
Microsoft Hesitate: https://www.theinformation.com/articles/why-sam-altman-joined-forces-with-larry-ellison-and-took-a-step-back-from-microsoft?rc=sy0ihq

Dylan Patel o3+ for Anthropic: https://www.youtube.com/watch?v=7EH0VjM3dTk


Deepseek R1: https://arxiv.org/pdf/2501.12948
https://arxiv.org/pdf/2412.19437
Diagram: https://pbs.twimg.com/media/GhyQsM6WQAE7W52?format=jpg&name=large
https://simple-bench.com/

Process: https://x.com/sama/status/1664018190840614912
https://openai.com/index/improving-mathematical-reasoning-with-process-supervision/
https://x.com/karpathy/status/1835561952258723930
https://openai.com/index/trading-inference-time-compute-for-adversarial-robustness/?s=09

Demis Interview: https://www.youtube.com/watch?v=yr0GiSgUvPU

Humanity’s Last Exam:
https://agi.safe.ai/
https://x.com/DanHendrycks/status/1882481730671857815
https://www.nytimes.com/2025/01/23/technology/ai-test-humanitys-last-exam.html?s=09


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

Podcast: https://aiexplainedopodcast.buzzsprout.com/