Hey HN. You probably saw the ChatGPT Space launch at OpenAI's DevDay. No shade on the product or the team, but I think it represents something that could get very scary: centralised control of our context.

I wanted to share some thoughts, and a two minute demo video of an open-source alternative I'm developing. Surely, our context should belong to us?

How does antitrust law handle when a company's technology increasingly approaches the general ability to do the job of every other technology? My guess is not very well.

This feels like a very logical commercial move from the labs but one that is really worrying, if successful. As open-weights and other models catch up with the capability, I can understand why they would try to lock people in another way. But the idea of having all your context locked behind a paywall is pretty scary!