1,307 karma · joined September 25, 2024
Shouldn't they have a reason, or am I just not seeing it?
[0]: https://www.anthropic.com/research/end-subset-conversations
At least choosing a different theme is the easiest it's ever been. Unlike the rounding, no CSS hacking is necessary.
I would love to sue every SaaS company for hard-to-understand terms too, as you suggest, but at what point would Anthropic actually need to tell someone about a terrorist manifesto that their customer gave them to read... Never?
...But since they exist, the company acted more responsibly than if they had simply ignored the data they chose to observe. Which is both the bare minimum given the circumstances, but also strangely absent almost all the time in the industry (including within Anthropic).
I am not a lawyer, but I have the feeling that reporting a discovered terrorist manifesto or something similar is in the spirit of the law.
> Review is needed to enforce our Usage Policy... designated members of our Trust & Safety team may access this data on a need-to-know basis as a part of their evaluation process.
[0]: https://privacy.claude.com/en/articles/10458704-how-does-ant...
if(q=="can you feel pain") print "yes"The most poignant criticism of the use of the algorithm itself happens almost accidentally, at the end of the article, by the tech executive being interviewed.
I asked what Claude would think of the pope’s encyclical. It was, after all, the world’s most significant moral document to date on A.I. Mr. Olah hesitated. “Things that go on the internet do affect models,” he said, visibly uncomfortable. But Anthropic would not specifically use the document to further train Claude, he said. The strongest influence on Claude would be Anthropic’s own training.
[0]: https://www.theguardian.com/technology/2026/mar/01/claude-an...
e.g. https://www.firefox.com/en-US/firefox/152.0.4/releasenotes/
[0]: https://en.wikipedia.org/wiki/Casualties_of_the_Gaza_war
https://apnews.com/article/meta-ai-hacking-anthropic-irregul...
You can find many more examples by searching major media outlets for words like rogue AI.
When I read the title, my initial thought was "did someone besides OpenAI use their product?" Then I opened the article to find out OpenAI was responsible.
Granted, my phone keyboard lacks a substantial dictionary, offers words instead of tokens, has a very small context window, and doesn't randomize outputs (Ted Bungie is its primary recommendation every time), but those are basically parameters to the existing autocomplete.
A machine running a loop through an expensive LLM for an undisclosed amount of time, which cost an undisclosed amount of money, which was told to keep looping into a solution was found, for a problem that nobody was very concerned about... That just seems like PR, and it's not so interesting.
> Smart Window infers information about you based on your interactions with the assistant and your Firefox browsing activity, including websites you have visited. This may include activity outside of Smart Window (such as in classic windows)...
It's a response to the Mozilla CEO saying[0] "Trust is earned. Consent is asked. AI skipped both," which promised to ameliorate this issue, and apparently did not.
> Previously, access to these changes required you to share technical and interraction data (telemetry) with Mozilla. However, as of Firefox version 148, you can opt into receiving remote improvements, even if you have opted out of sharing your telemetry or participating in our experimental studies.
Did they? There's a lot of implication with that "if", and I saw someone else point out (albeit rudely) that anything that may have supported them was taken down.
[0]: https://www.ghacks.net/2025/02/27/mozillas-new-terms-of-use-...