The AI feedback loop: Researchers warn of 'model collapse' as AI trains on AI-generated content
As a generative AI training model is exposed to more AI-generated data, it performs worse, producing more errors, leading to model collapse.
Are better models better? — Benedict Evans
Every week there’s a better AI model that gives better answers. But a lot of questions don’t have better answers, only ‘right’ answers, and these models can’t do that. So what does ‘better’ mean, how do we manage these things, and should we change what we expect from computers?
I don’t think we are close to “AI scientists”
Today's AI agents are not designed to extract deep insights from new observations.
Anthropic blames dystopian sci-fi for training AI models to act “evil”
But training on "synthetic stories" that model good AI behavior can help.
Unfortunately, You Need to Know What the Jevons Paradox is
Unfortunately, You Need to Know What the Jevons Paradox is
How Americans see AI impacting human skills; its role in science, matchmaking, religion and more
Majorities don’t want AI to give advice about religion or love. But many say AI should help forecast weather, search for financial crimes, develop medicine and more.
AI is doing something weird to Science
In both good and bad ways, and it won't go away.
Look-ahead Reasoning with a Learned Model in Imperfect Information Games
Test-time reasoning significantly enhances pre-trained AI agents' performance. However, it requires an explicit environment model, often unavailable or overly complex in real-world scenarios....
On AI as model organism for human learning
I've a pretty taste for paradox
AI Has Ruined the Job Market
Maybe flawed people were better than brute algorithms.
Experts warn AI is making your brain work less
Generative AI tools have become hugely popular but some experts worry about the effect they have on the brain.
AI is getting too advanced
AI Is a Waste of Time
The newest AI tools are accelerating basic research and scaring the general public. But many people are simply using them as toys.
If we just make reasonable comparisons instead of crazy ones
New research: how well do AI models actually follow their constitutions? 205 tenets from Anthropic's 30K-word soul doc. Adversarial multi-turn scenarios against 7 models. Claude: 15% → 2% violation rate in two generations. Training works. But the remaining failures tell a more important story.
