When language models “reason out loud,” it’s hard to know if their stated reasoning is faithful to the process the model actually used to make its prediction. In two new papers, we measure and improve the faithfulness of language models’ stated reasoning.
SAFETY
-
AI War and Global Arms Race: TED Talk Discussion
By
–
I recently spoke on this matter for a @TEDTalks
, on AI, War, and the new Global Arms Race. -
AI Worst-Case Scenario: Ethics and Digital Transformation
By
–
#AI: The worst-case scenario https://
bit.ly/46ldAUY #ethics #DigitalTransformation -
Language Conditioned Traffic Generation for Autonomous Driving Simulation
By
–
Language Conditioned Traffic Generation paper page: https://
huggingface.co/papers/2307.07
947
… Simulation forms the backbone of modern self-driving development. Simulators help develop, test, and improve driving systems without putting humans, vehicles, or their environment at risk. However, -
Harari discusses AI dangers, human nature and civilization origins
By
–
Here's my conversation with Yuval Noah Harari (
@harari_yuval
) about human nature, intelligence, power, war, communism, fascism, origins of human civilization & the dangers of AI. Yuval also responds critically to my conversation with Benjamin Netanyahu. https://
youtube.com/watch?v=Mde2q7
GFCrw
… -
Integrating Trustworthy AI: OSTP Bill of Rights and NIST Framework
By
–
Interested in how to integrate #trustworthyAI into the AI lifecycle? Check out Global AI Ethicist Haniyeh Mahmoudian in an on-demand webinar, where she discusses the #OSTP AI Bill of Rights and #NIST AI Risk Management Framework, which addresses trust, safety, and ethical
-
Modern Turing Test: Measuring AI Progress Toward Artificial Capable Intelligence
By
–
AIs today are close to mastering language… We need a new milestone – a Modern Turing Test – to measure progress for the next phase. Once they can take actions, we have artificial capable intelligence (ACI) and the full transition to a post-AI world begins. What matters is
-
Autonomous Vehicles Responding to Emergency Vehicles in San Francisco
By
–
Sunday 10am 20 blocks across PacHghts SF, saw 3 @Waymo
, 4 @Cruise humanless. Latter pulled over to curb, stopped, and emergency lights when oncoming fire truck w/ flashing lights. First such sighting for me. Well done! Any one seen behavior when fire coming from behind? -
Debunking Five Common AI Myths and Misconceptions
By
–
Debunking AI Myths: The Truth Behind 5 Common #Misconceptions Unravel the enigma of #AI as this article busts five pervasive #myths, from #fears of robotic world domination to the mistaken belief in #machines' #unbiased judgment.
-
New Book Explores AI and Climate Challenge Future
By
–
Au cœur de la Provence, j'avance avec passion sur la rédaction de mon nouveau livre « Le Nouveau Temps ». Prévu pour cet automne, il sera une immersion dans le futur à l'heure de l'IA et du défi climatique. Hâte de partager ces réflexions avec vous ! #LeNouveauTemps