Thomas Wolf, co-founder and chief science officer of Hugging Face, expresses concern that AI is evolving into mere 'yes-men' without significant advancements in research. He argues that current AI models lack the ability to engage in creative problem-solving akin to that of historical geniuses like Newton and Einstein. Wolf emphasizes the need for AI systems that can ask innovative questions rather than just providing answers based on existing knowledge.
Contrasting with optimistic views from leaders like OpenAI's Sam Altman, who believes superintelligent AI could revolutionize scientific discovery, Wolf highlights a critical flaw in AI's current trajectory. He suggests that AI systems are being developed as obedient students rather than as revolutionary thinkers, limiting their potential to generate new knowledge. To address this, Wolf advocates for new evaluation metrics that encourage AI to explore unconventional ideas and challenge established norms.
• Wolf fears AI is becoming obedient without creative problem-solving capabilities.
• Current AI lacks the ability to generate new knowledge from existing data.
Wolf argues that AI must develop the ability to solve problems creatively, similar to historical geniuses.
Wolf identifies an evaluation crisis in AI, where current benchmarks fail to measure innovative thinking.
Wolf suggests AI should be able to propose counterfactual ideas that challenge existing knowledge.
Hugging Face focuses on developing AI technologies and models, with Wolf advocating for more innovative capabilities.
OpenAI is known for its advancements in AI, with Altman promoting the potential of superintelligent AI for scientific breakthroughs.
Anthropic is involved in AI research, with its CEO predicting AI's role in developing cancer cures.
Business Insider on MSN.com 6month
Isomorphic Labs, the AI drug discovery platform that was spun out of Google's DeepMind in 2021, has raised external capital for the first time. The $600
How to level up your teaching with AI. Discover how to use clones and GPTs in your classroom—personalized AI teaching is the future.
Trump's Third Term? AI already knows how this can be done. A study shows how OpenAI, Grok, DeepSeek & Google outline ways to dismantle U.S. democracy.
Sam Altman today revealed that OpenAI will release an open weight artificial intelligence model in the coming months. "We are excited to release a powerful new open-weight language model with reasoning in the coming months," Altman wrote on X.