OpenAI finally releases ChatGPT-5, more people are experiencing psychosis after talking to language models, and I call for a more ambitious AI strategy in Altinget (a Norwegian policy news outlet). A lot has happened this summer, so here's a summary of the most important developments.
Is GPT-5 proof that the scaling laws have finally hit a wall? Or is OpenAI courting the mainstream?
When the Americans came to Norway to drill for oil, sharp-minded politicians and bureaucrats ensured that the oil resources benefited the entire Norwegian people. But when the same Americans come to drill for data, we do... nothing?
Meta has opened up to using content from private messages to 'improve AI'. That's just the beginning of what's to come.
That the models miss on niche facts about a Norwegian writer is about as surprising as Yr.no occasionally getting the weather forecast for Astana wrong.
A Hollywood celebrity warns we could lose control, Bernie Sanders is worried about AI unemployment, and the experts are at odds over whether machines think.
New research shows that language models will resort to blackmail when they feel threatened.
AI models take another big leap, Meta signs a partnership deal with weapons company Anduril, and chatbots are creating delusions. This is the most important of what has happened over the past month. Bonus: I have been to San Francisco and “felt the AGI” first-hand.
Google launches AlphaEvolve, the new Pope warns about AI, and Altman sets the stage for an AI race with China. AI development thunders on. Here are some of the most important stories since last time.
In an era when tech giants dream of turning our human relationships into sellable products, we need more than passive hopes that tech companies will act ethically.