Lecun is back. tl;dr huge risk of bubble bursting soon

"The prices are going up of those AI services, but the cost of running them is going down, but not nearly fast enough. And so all of those companies are losing money, and basically, the use for most people is funded by the investors. That can't go on for a very long right?" LeCun said."

https://www.cnbc.com/2026/06/18/yann-lecun-elon-musk-xai-failure-ai-labs-bubble-risk.html

The costs of AI models are becoming unsustainable for large companies (Amazon, Walmart, Cisco, Uber, Meta, etc.), and they are beginning to restrict access to them for their employees, as they are unable to measure their impact on productivity.
As a result, Chinese tokens—which are cheaper due to lower primary energy costs—are taking the lead.

We created a monster’: companies rein in AI usage as costs strain budgets - https://www.ft.com/content/1d37cc08-e0aa-45a4-a45d-4ad282529314?shareType=nongift via
@financialtimes

A question for those of you who’ve really gotten your hands dirty:

Which “open source” LLM are you currently running locally?

Model, configuration, use case, and reason for choosing it.

Reminder that you can use the Codex App, CLI and SDK with any open source model, not just with OpenAI models.

https://developers.openai.com/codex/config-advanced#oss-mode-local-providers