The article itself is so lacking in substance that it’s hard to even say it’s misleading. The part that seems to be about “AI is too expensive” reads:
Fortune, citing The Verge, said that Microsoft steered engineers away from Anthropic’s Claude Code and over to GitHub Copilot CLI, even though access to Claude Code was opened only about six months ago.
Which isn’t “AI is too expensive”, it’s “our in-house AI was cheaper than Anthropic’s service.”
And the whole rest of the article is just the usual vague “not everyone finds AI useful for everything” and “water usage? Power grids?” And “by 2030 there’ll be a lot more tokens used than today” (which seems contrary to the headline, but whatever).
The article sucks. That doesn’t change that most companies are still not even remotely cost controlling LLM usage like pretty much anyting else, nor that we even know what real costs, without billions in venture capital subsidies, would look like, other than substantially higher than now.
The article itself is so lacking in substance that it’s hard to even say it’s misleading. The part that seems to be about “AI is too expensive” reads:
Which isn’t “AI is too expensive”, it’s “our in-house AI was cheaper than Anthropic’s service.”
And the whole rest of the article is just the usual vague “not everyone finds AI useful for everything” and “water usage? Power grids?” And “by 2030 there’ll be a lot more tokens used than today” (which seems contrary to the headline, but whatever).
The article sucks. That doesn’t change that most companies are still not even remotely cost controlling LLM usage like pretty much anyting else, nor that we even know what real costs, without billions in venture capital subsidies, would look like, other than substantially higher than now.