Google restricting Google AI Pro/Ultra subscribers for using OpenClaw

Post by sowbug » Mon, Feb 23, 2026, 2:01 AM UTC

lysace wrote: post #61

avazhi wrote: post #27

Google deciding to willy nilly unilaterally ban my 20+ year old primary Google account is probably my greatest internet fear, given how famously awful their support is. Seems like it's the singular best example of a tech company so big that through some combination of internal silos and TOS bureaucracy you have no shot of getting your account back, no matter how unreasonable the ban actually is. A while back I made c…

To clarify: None of the comments in that thread talk about experiencing that. They have been locked out of the Gemini service, not their Google account with mail etc. Source: I actually read them. Yes, personally. I didn't even have an LLM summarize them. I know, I'm practically a luddite.

Post by easton » Mon, Feb 23, 2026, 2:01 AM UTC

martinald wrote: post #125

obblekk wrote: post #111

This is the first time in recent memory that software has had high variable costs so the surprise at these rules is understandable. In this case, a the difference in context cache hit rate between openclaw and antigravity. For example if openclaw starts every message with the current time hh:mm:ss at the top of the context window, followed by the full convo history, it would have a cache hit rate if ~0. Simply moving…

It's a fair point, but I think people are thinking too much about 'cost' and 'subsidies' and just the fact that everyone is so compute stretched. While it's sort of the same thing, I think it's much more a symptom of not enough compute vs some 'dump cheap tokens' on the market strategy. One related thought I had was that given OpenAI is the only one _not_ doing this of the big3, it probably indicates they have a lot…

Post by nikcub » Mon, Feb 23, 2026, 2:01 AM UTC

mvdtnz wrote: post #190

nikcub wrote: post #182

Earlier quoted context omitted.

> they say they use heuristics and usage patterns. cache hit rate alone would stand out

Why do you mean by this? What cache?

Post by overfeed » Mon, Feb 23, 2026, 2:06 AM UTC

sowbug wrote: post #109

noosphr wrote: post #45

Earlier quoted context omitted.

You are being subsidised to the tune of 50 to 99.9 cents on the dollar compared to the API. What the hell do you expect? To get paid for using other people's tools on Google's servers?

Businesses do not have an entitlement to profit. Suspending customers for using a fairly expensive subscription plan -- especially forfeiting an annual prepayment for a day or two of coloring outside the lines -- sure does make Google appear entitled to profit without ever risking its own pricing model.

Post by adastra22 » Mon, Feb 23, 2026, 2:09 AM UTC

merlindru wrote: post #169

gck1 wrote: post #149

I don't understand how this can be enforced without ridiculous levels of false positives. I'm truly baffled. The same with Claude Code situation. gemini-cli, claude-code, codex etc, they ALL have a -p flag or equivalent, which is non-interactive IO interface for their LLM inference. If I wire my tooling (or openclaw) to use the -p flag (or equivalents), is that allowed? Okay, maybe they get rid of the -p flag and I h…

claude -p is allowed as far as I'm aware. if i understand correctly, they even have a wrapper around it to make it easier to use: the Claude Agent SDK the thing that's disallowed is pretending you're the claude binary, logging in through OAuth in other words, if you use some product thats not Claude Code, and your browser opens asking you to "give Claude Code access to your account", you're in hot water as for how th…