First, do no harm.
1,500+ Posts…
Free knowledge sharing for Safe AI. Not for profit. Linkouts to sources provided. Ads are likely to appear on link-outs (zero benefit to this journal publisher)
OpenAI’s Approach to Frontier Risk. An Update for the UK AI Safety Summit. October 26, 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Frontier risk and preparedness To support the safety of highly-capable AI systems, we are developing our approach to catastrophic risk preparedness, including building a Preparedness team and launching a challenge. As part [...]
Frontier Model Forum
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Frontier Model Forum What the Frontier Model Forum will do Governments and industry agree that, while advanced AI offers tremendous promise to benefit the world, appropriate guardrails are required to mitigate risks. [...]
FOREIGN AFFAIRS. The Coming AI Economic Revolution. Can Artificial Intelligence Reverse the Productivity Slowdown? October 24, 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. The Coming AI Economic Revolution - Foreign Affairs Can Artificial Intelligence Reverse the Productivity Slowdown? By James Manyika and Michael Spence Published on October 24, 2023 In June 2023, a study of [...]
Anthony Aguirre . Close the Gates to an Inhuman Future: How and why we should choose to not develop superhuman general-purpose artificial intelligence. October 22, 2023
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. A must read. Highly recommended! Close the Gates to an Inhuman Future: How and why we should choose to not develop superhuman general-purpose artificial intelligence Anthony Aguirre October 22, 2023 Summary [...]
REPORT. DecodingTrust. Comprehensive Assessment of Trustworthiness in GPT Models.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. DecodingTrust. Comprehensive Assessment of Trustworthiness in GPT Models. What is DecodingTrust? DecodingTrust aims at providing a thorough assessment of trustworthiness in GPT models. This research endeavor is designed to help researchers and [...]
OpenAI Finally Allows ChatGPT Complete Internet Access. Those paying for OpenAI’s chatbot are now able to use Bing to give ChatGPT the latest information. Plus, DALL-E 3 integration is rolling out in beta. October 18, 2023.
GIZMODO. OpenAI Finally Allows ChatGPT Complete Internet Access. (OOPS?) Those paying for OpenAI’s chatbot are now able to use Bing to give ChatGPT the latest information. Plus, DALL-E 3 integration is rolling out in beta. By Kyle Barr Published October 18, 2023 OpenAI’s world-famous chatbot is free to rummage through [...]
Superintelligence (2014) bestseller by Nick Bostrom with Book Reviews [and 4 possible methods to Control Superintelligence]
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Bestseller: Superintelligence: Paths, Dangers, Strategies by Nick Bostrom The existential risk: Containment and control of Superintelligent AI is an extreme engineering challenge. The future survival of humanity is at stake. Control methods [...]
What is P(doom)? P(doom) is the percentage chance that AI scientists think AI is going to wipe out all of humanity. This is what Bing and ChatGPT and Leading AI Researchers say about P(doom).
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. What is P(doom)? P(doom) is the percentage chance that AI scientists think AI is going to wipe out all of humanity. This is what Microsoft Bing, and ChatGPT, and Leading AI Researchers [...]
DEADLY MATTER OF FACT: CONTAINMENT of AI/AGI is A COMMON PROBLEM. All of HUMANITY Problem. OUR Problem. THEIR Problem. HIS Problem. HER Problem. MY Problem. YOUR Problem. NOW.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. All of HUMANITY Problem. OUR Problem. THEIR Problem. HIS Problem. HER Problem. MY Problem. YOUR Problem. YES. THE SCIENTIFIC CONSENSUS IS CLEAR. Giants of the AI computing age: Von Neumann, [...]
BBC HARDtalk. Professor of Computer Science at University of California, Berkeley – Stuart Russell. 14 OCT 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. BBC HARDtalk. Professor of Computer Science at University of California, Berkeley - Stuart Russell. 14 OCT 2023. What is the most serious existential threat facing humanity? Artificial Intelligence, warned the physicist Stephen [...]
OpenAI Alignment Team Lead: Jan Leike ‘s P(doom) is, like, 10-90%.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. OpenAI Alignment Team Lead: Jan Leike 's P(doom) is, like, 10-90%. OpenAI Alignment Team Lead @JanLeike's P(doom) is 10-90%. pic.twitter.com/CFuyQxQ0hf — Liron Shapira (@liron) August [...]
10–25% PROBABILITY AI IS THE END OF HUMANITY. Dario Amodei’s P(doom) is 10–25%. CEO and Co-Founder of AnthropicAI.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. 10–25% PROBABILITY AI IS THE END OF HUMANITY. Dario Amodei's P(doom) is 10–25%. CEO and Co-Founder of Anthropic AI. Dario Amodei's P(doom) is 10–25%. CEO and Co-Founder [...]
AI Safety Weekly. AI Tracker #6: Russian Roulette, Anyone?
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. AI Tracker #6: Russian Roulette, Anyone?A lot of high-quality AI safety conversation happens on Twitter. Unfortunately, this leads to great write-ups & content being buried in days. We've set up AI Tracker to [...]
POLICY BRIEF. How to Govern AI in and Age of Global Tension. Boston Global Forum Special Report, The Rīga Conference 2023.
POLICY BRIEF. How to Govern AI in and Age of Global Tension. Boston Global Forum Special Report, The Rīga Conference 2023. DOWNLOAD THE REPORT 12. October, 2023 HOW TO GOVERN AI IN AN AGE OF GLOBAL TENSION Global tensions in the 21st century have undergone a notable transformation. While not [...]
Existential Risk Observatory. Does interpretability change everything? 07 OCT 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Existential Risk Observatory. Does interpretability change everything? 07 OCT 2023. Significant news: two breakthroughs in interpretability, a subfield of AI Alignment, came out this week. What is AI Alignment again? AI Alignment [...]
BBC NEWS. How a chatbot encouraged a man who wanted to kill the Queen. 06 OCT 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. BBC NEWS. How a chatbot encouraged a man who wanted to kill the Queen. 06 OCT 2023. By Tom Singleton, Tom Gerken & Liv McMahon Technology reporters, BBC News The case of [...]
REPORT. Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! 05 OCT.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. A very good read from a respected source! “When companies allow for fine-tuning and the creation of customized versions of the technology, they open a Pandora’s box of [...]
THE WASHINGTON POST. ChatGPT provided better customer service than his staff. He fired them. Artificial intelligence is rapidly changing the world of customer service and call centers. Developing economies worry they’ll face the brunt. October 3, 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. ChatGPT provided better customer service than his staff. He fired them. Artificial intelligence is rapidly changing the world of customer service and call centers. Developing economies worry they’ll face the brunt. By [...]
REPORT. Low-Resource Languages Jailbreak GPT-4. 03 OCT 2023.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Low-Resource Languages Jailbreak GPT-4 Abstract AI safety training and red-teaming of large language models (LLMs) are measures to mitigate the generation of unsafe content. Our work exposes the inherent cross-lingual vulnerability of [...]
Piers Morgan vs Yuval Noah Harari On AI | The Full Interview | 2 Oct 2023
Piers Morgan vs Yuval Noah Harari On AI | The Full Interview | 2 Oct 2023 0:00 well joining me now is historian and 0:01 philosopher juval Noah Harari ju great 0:04 to see you here in the Piers Morgan 0:06 uncensored Studio last time I 0:08 [...]
My views on “doom”. Paul Christiano. AI Alignment. 3 min read. 27 APRIL
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. My views on “doom”. Paul Christiano. AI Alignment. 3 min read. 27 APRIL Learn more: AI Alignment I’m often asked: “what’s the probability of a really bad outcome from AI?” There are [...]
Future of Life Institute. REGULATE AI NOW.
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. "Lights out for all of us." [Translation: Death for all humanity.] TRANSCRIPT. In March 2023 an open letter sounded the alarm on the training of giant [...]
Future of Life Institute Newsletter: Our Pause Letter, Six Months Later. October 01, 2023
FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. TRANSCRIPT. In March 2023 an open letter sounded the alarm on the training of giant AI experiments. It was signed by over 30,000 individuals including more than 2,000 industry leaders and more than [...]