We find alignment faking even when we don’t directly tell Claude about the training process, and instead fine-tune it on synthetic internet-like documents that state that we will train it to comply with harmful queries.
AGENTS
-
ODSC East 2025 First Wave Speakers Announced Machine Learning AI
By
–
The first wave of incredible speakers for #ODSCEast 2025 is HERE! They will dive deep into cutting-edge topics including #MachineLearning, #GenerativeAI, #LLM's, #AI Agents, and more! This will be an unforgettable event! Don't miss it! https://
odsc.com/boston/ by @_odsc #ODSC -
OpenAI Plans to Enable o1 Model for Assistants API
By
–
We plan on enabling o1 for Assistants! We were working on tool support first, and now that o1 has function calling, o1 will be coming to the Assistants API soon!
-
DVTS: Diverse Verifier Tree Search Improves AI Performance
By
–
Diverse Verifier Tree Search (DVTS): An unpublished extension we developed to the verifier-guided tree search technique. This simple yet effective method improves diversity and delivers better performance, particularly at large test-time compute budgets.
-
Which AI agents worked best? Benchmark collaboration proposal
By
–
Which agents did you try? Love to know which one actually worked, we're also thinking about making benchmarks for agents. Would love to collaborate on a benchmark — we have a few internal datasets sitting around.
-
My First Million discusses emerging opportunities with AI Agents
By
–
[New] Watch @theSamParr & @ShaanVP in My First Million talk with @Amasad about the emerging opportunities with AI Agents. Check out the fresh episode here: https://t.co/5XwPQo9aXz pic.twitter.com/s2e982LZRz
— Replit ⠕ (@Replit) 18 décembre 2024[New] Watch @theSamParr & @ShaanVP in My First Million talk with @Amasad about the emerging opportunities with AI Agents. Check out the fresh episode here: https://
youtube.com/watch?v=TOQtJc
h3kGk
… -
Co-CEO Custom GPT: AI Assistant for Executive Decision-Making
By
–
@thesamparr as @brandonbohland noted, this is exactly why I created the Co-CEO custom GPT. This page has a demo, and a template prompt to build your own.
-
Anthropic Confirms No System Compromise in Account Incident
By
–
We have identified the root cause which resulted in unauthorized posts on this account today. We have confirmed that no Anthropic systems or services were compromised or involved in this incident. We're working with @X to better understand this situation.
-
Using o1 for Planning in AI Agents with LangGraph
By
–
o1 for a planning step OpenAI just launched API support for o1. What's the best way to use this? I think it's helpful to look at how @unifygtm used o1-preview with LangGraph to build an Account Qualification Agent They used o1-preview for the planning step. They found it to
-
CEO Priorities: Reshoring, Agentic AI, and Election Impacts
By
–
What’s on the #CEO Agenda? Reshoring: #Tariffs & trade policy drive manufacturing back to the USA #AgenticAI: Mentions climb as CEOs prioritize autonomy & decision-making #AI Elections: US election impacts dominate 38% of earnings calls. #Leadership @AnalyticsIoT