Artificial intelligence is accelerating – but how fast is too fast? A new benchmark from research group METR suggests that the latest AI models from Anthropic, OpenAI and Google are improving at an exponential rate in real-world coding tasks. Some researchers now warn that AI progress could be far more disruptive than most people realise. If models like GPT-5, Claude Opus 4.6, and Gemini 3 Pro are getting dramatically better at software development, are we approaching a technological tipping point? Could AI begin accelerating its own research? And what does this mean for jobs in tech and other sectors? Our technology correspondent Rowland Manthorpe explores whether we are witnessing a breakthrough moment in artificial intelligence and what happens next.
It's Just Technology