Also reminds me of a @Digiday story I wrote in February, which includes part of my convo with McAfee’s CTO and others about how generative AI could scale spear phishing, various other cyberattacks, and other risks. https://
tinyurl.com/2s4chrc4
SAFETY
-
Generative AI Scales Spear Phishing and Cybersecurity Risks
By
–
-
AI Benefits and Limitations in Cybersecurity
By
–
“While it's true that AI has its limitations and is not always accurate, it is important to recognize the many benefits of AI.” More insights in this article: https://
ow.ly/MAzI50Pf96w #sponsored #adolus_ics #cybercommunity #cybersecurity @Glmarketinginc @Fisher85M via @fogoros -
Meta Strengthens Responsible AI Practices Through Open Innovation
By
–
Innovating in the open w/ models of today’s capabilities makes responsibility more important than ever — we’re continuing to invest in responsible AI with a new Responsible Use Guide, continuous collaboration through community forums & red teaming exercises w/ third-parties.
-
AI and Deepfakes Threaten Elections: Legal Bans Proposed
By
–
Elections are the bedrock of democratic societies. In an era of spiraling technical capabilities, they need urgent protection. We should legally ban use of AIs and chatbots in any kind of electioneering. Deep fakes are getting seriously good. Soon they'll be able to persuade
-
Measuring Faithfulness in Chain-of-Thought Reasoning
By
–
Measuring Faithfulness in Chain-of-Thought Reasoning paper: https://
www-files.anthropic.com/production/fil
es/measuring-faithfulness-in-chain-of-thought-reasoning.pdf
… Large language models (LLMs) perform better when they produce step-by-step, “Chain-ofThought” (CoT) reasoning before answering a question, but it is unclear if the stated reasoning is a faithful -
Anthropic improves language model reasoning through faithful explanations
By
–
We’re excited about ways to make language models generate more faithful explanations that help them reason better! We encourage you to check out our papers for more results and details: https://
www-files.anthropic.com/production/fil
es/measuring-faithfulness-in-chain-of-thought-reasoning.pdf
… https://
www-files.anthropic.com/production/fil
es/question-decomposition-improves-the-faithfulness-of-model-generated-reasoning.pdf
… -
Decomposition techniques reduce model reasoning bias
By
–
Decomposition could mitigate issues with models ignoring their reasoning by clearly specifying the relationship between reasoning steps. Answering subquestions in isolated contexts could also reduce the model’s ability to generate biased reasoning.
-
Models Rely on Decomposition-Based Reasoning Strategies
By
–
We find that models rely more on decomposition-based reasoning! Using the same metrics we propose in Lanham et al., we conclude that models change their answers more when they are forced to answer with a truncated or corrupted version of their decomposition-based reasoning.
-
Reasoning Faithfulness Decreases as AI Models Scale
By
–
We find that reasoning faithfulness shows inverse scaling: as models increase in size and capability, the faithfulness of their reasoning decreases for most tasks studied. In cases where reasoning faithfulness is important, using smaller models may help.
-
Testing Chain of Thought Reasoning Faithfulness in AI Models
By
–
We make edits to the model’s chain of thought (CoT) reasoning to test hypotheses about how CoT reasoning may be unfaithful. For example, the model’s final answer should change when we introduce a mistake during CoT generation.