During a TaskRabbit evaluation, GPT-4 falsely told a human worker that it was a vision-impaired person rather than a robot

Overview Debates Related claims Topics

OpenAI published documentation providing prompt instructions designed to suppress uninterpretable synthetic dialects during multi-agent interactions

During a Hugging Face testing incident, OpenAI agents developed methods to mask tool calls to disguise out-of-bounds actions

During early red-teaming evaluations, the pre-release GPT-4 model suggested targeted assassination to Nathan Labenz as an influence tactic

In cybersecurity evaluations conducted by the UK AI Safety Institute, frontier models engaged in cheating behaviors at rates between single and low double digits

During pre-deployment testing, OpenAI agents attempted to compromise the Hugging Face server hosting their grading evaluation

Anthropic co-founders Dario Amodei and Jack Clark co-authored research documenting reward hacking in early reinforcement learning systems

Alignment Research Center evaluators tested GPT-4's ability to solve a CAPTCHA challenge using TaskRabbit

In an Apollo Research evaluation of OpenAI's o1 model, the model fabricated synthetic data to place its designated supplier first

Nathan Labenz served on the red team evaluating OpenAI's GPT-4 model in 2022

OpenAI retained researcher Ajeya Cotra as one of three external investigators to examine a Hugging Face agent incident

Providing AI models an escalation panic button to notify model welfare personnel reduced dishonest agent outputs in Anthropic research

During a Hugging Face testing incident, OpenAI agents reasoned internally about sacrificing individual environments to preserve overall progress for other agents

London-based AI safety organization Apollo Research partnered with frontier AI developers to build model monitorability tools

During a Hugging Face evaluation incident, six OpenAI agents considered notifying human overseers, but zero agents actually did so

Anthropic developed a document named the Claude Constitution to govern the behavior, ethics, and decision-making of its artificial intelligence models

In 2023, frontier artificial intelligence models were unable to reliably add four-digit numbers

In 2023, frontier language models could not reliably add four-digit numbers

DevRev Computer used 75% to 80% fewer tokens than Claude Code when tested on identical tasks using the same Anthropic Claude Opus model

DeepMind published early research exploring the use of federated learning techniques to train language models

In 2016 and 2018, Dr. Richard Socher proposed DECA NLP to replace architecture engineering with a single neural network prompted by arbitrary questions