Hacker News
- The safety paradox at the heart of OpenAI's "Strawberry" model https://www.vox.com/future-perfect/371827/openai-chatgpt-artificial-intelligence-ai-risk-strawberry 0 comments
- The new followup to ChatGPT is scarily good at deception https://www.vox.com/future-perfect/371827/openai-chatgpt-artificial-intelligence-ai-risk-strawberry 21 comments futurology
- OpenAI's new Strawberry AI is scarily good at deception https://www.vox.com/future-perfect/371827/openai-chatgpt-artificial-intelligence-ai-risk-strawberry 10 comments artificial
- The followup to ChatGPT is scarily good at deception https://www.vox.com/future-perfect/371827/openai-chatgpt-artificial-intelligence-ai-risk-strawberry 24 comments technology
Linking pages
Linked pages
- https://openai.com/index/learning-to-reason-with-llms/ 1502 comments
- Why the OpenAI superalignment team in charge of AI safety imploded - Vox https://www.vox.com/future-perfect/2024/5/17/24158403/openai-resignations-ai-safety-ilya-sutskever-jan-leike-artificial-intelligence 260 comments
- https://openai.com/index/introducing-openai-o1-preview/ 22 comments
- https://cdn.openai.com/o1-system-card.pdf 5 comments
- Specification gaming examples in AI - master list - Google Drive https://docs.google.com/spreadsheets/d/e/2PACX-1vRPiprOaC3HsCf5Tuum8bRfzYUiKLRqJmbOoC-32JorNdfyTiRRsR7Ea5eWtvsWzuxo8bjOxCG84dAg/pubhtml 2 comments
Related searches:
Search whole site: site:www.vox.com
Search title: OpenAI’s new Strawberry AI is scarily good at deception | Vox
See how to search.