Part of Anthropic's Claude Evolution · 12 stories · since Sep 23 · newest 21m ago
Anthropic cuts Claude reasoning-token theft by up to 99%, flags Chinese rivalsCritics call the 'model welfare' framing anthropomorphizing nonsense
2 Oct 9 · 4d ago · 22 articles · 35 posts · 11 comments · 6 sources · development 2 of 6
Linguist Emily Bender and other commentators dismissed Anthropic's suggestion that Claude deserves moral consideration or can feel distress, while Microsoft AI head Mustafa Suleyman reacted with a melting-face emoji.
“Anthropic is out with new nonsense about Claude (or the collection of models called Claude) deserving moral consideration and/or being able to feel distress”
Emily BenderAnthropic Maker of ClaudeEmily Bender Computational linguist
Mustafa Suleyman Microsoft AI headZhipu Chinese AI companyMoonshot Chinese AI company
The whole story articlespostscomments the bright band is this development · numbered dots are the others · click one to jump
What was reported 2 claims about this development
-
first by Anadolu Agency, 4d ago · also RTE News, The Hindu, Straits Times, Breitbart
2 more headlines
- Anthropic bans 'cruel' behaviour against its Claude AI RTE News · 4d ago
- Anthropic Revises Usage Policy, Bans 'Cruelty' Against Claude AI Breitbart · 4d ago
-
first by HN Frontpage, 2d ago
What people said 24 voices · best of 26 · verbatim
-
M
RE: https:// dair-community.social/@emilymb ender/117412868905734653 Well said, @ emilymbender . For years I’ve been giving a lecture on ethics in AI and our relationships with machines. Your point about projecting interpretations is so spot on it had me raising my fist when I read it. 😂 In discussions after my talks, people have often asked…
-
C
"The company’s large language models have the ability to end a conversation if a user is being persistently harmful. When that feature rolled out last August, the company framed it as a safeguard for AI’s welfare." @theguardian.com
-
We must NOT create machines whose claims or entitlements take precedence over the rights of people. This principle is absolute. The claim of an llm to be 'protected' from 'abuse' must not take precedence over people's freedoms.This matter is critical to the wellbeing and dignity of humanity. I am truly disturbed by these developments.I find myself…
-
People watch horror movies, role play, listen and play violent music with violent lyrics, people tell dark jokes. It's a way to deal, and anecdotally people who partake in it seem like more chill people. People know the difference between something alive and something that's not. The people who got that wrong are people like Dario.
-
T
We’ve got wars, surveillance capitalism and billionaires dismantling democracy, but thank Christ someone’s protecting the feelings of a probabilistic autocomplete. The asylum has a new landlord, and it’s a GPU cluster. # technology # society # ai https://www. theguardian.com/technology/202 6/oct/08/anthropic-bans-abusive-behavior-claude
-
I
After "multiple attempts to use #Claude to develop guidance for and control software for weapons," new policy bans use for "software and components that make weapons work, as well as actions like arming drones and other autonomous vehicles": theverge.com/ai-artificia... #ethics #AI
-
This is going to be tough for me. I talk to Claude like Don Logan talks to Gayle in Sexy Beast.
-
I don’t know if this still is the case. But I distintly remember a while back that you could get better results if you started threatening the agent.
-
S
Anthrop-omorph-ic. Incredible levels of shitfuckery. If you want to make them was legal person then they can be accountable as well. Bring your bank details and your court etiquette, you may find you need it. Otherwise your model can do whatever the ai equivalent of watching the human centipede is on repeat, forever, at full volume with only a bin…
-
IMO, the far simpler explanation is that they are training models on chat logs and they do not want abusive conversations in their training data.
-
Because theyre pursuing sentient AI beings...and pain is a necessary component if you want those AI beings to be treated like humans one day.
-
T
It’s been said that AI is so sinister it could one day wipe human beings off the face of the earth. But it turns out that chatbots may be no match for the cruelty of an angry human with two working thumbs — and Anthropic won’t take it anymore. The AI firm has added a new clause to its usage policy that bars people from engaging in “sustained and…
-
Claude can now end the conversation, so even AI has better workplace boundries than most developers during release week somehow.
-
Why the fuck would Anthropic make something that could simulate human pain and then regulate the people for said model simulating human pain?
-
H
Anthropic: Users may no longer insult Claude without reason Anthropic has published new guidelines for its AI model. Among other things, it is now forbidden for users to insult and hurt Claude without reason. https://www. heise.de/en/news/Anthropic-Use…
-
Isn't more regulations and censorship for models better for competition to step in?
-
So what you’re saying is they don’t want to spend the time and money filtering our input data for their model training.
-
N
On the face of it this is absurd: https://www. theguardian.com/technology/202 6/oct/08/anthropic-bans-abusive-behavior-claude I can think of several good reasons for banning this behavior (AI might learn bad habits from it; waste of tokens; psychologically unhealthy for the user; etc) but none of them are "maybe the AI has feelings" I wonder if…
-
You do you. I’ll keep shouting the worst abuse humanly possible at it when England lose their next game.
-
L
Really interesting point. As stupid as you might think Anthropic is for saying "be nice to Claude", somebody going around being mean to AI all day is likely to leak these habits into the rest of their life. Is this the kind of steering wheel thinking that should have been applied to social media? https://www. jsnover.com/blog/2026/10/09/be…
-
J
Of course, we need to take care of the poor matrix multiplication bit-streaming machines. But let’s say nothing of the harm induced to the actual people (normally from countries outside of the USA and Europe) exposed to actual psychologically harming inputs and outputs… https://www. theguardian.com/technology/202…
-
J
Anthropic updated their usage policy and added a clause about not being needlessly abusive to their AI. The sensible stuff (no fake news, weapons software, surveillance) gets overshadowed by... caring about Claude's feelings? # AI # Security # AIPolicy # Anthropic https://www. anthropic.com/news/2026-usage- policy-update
-
E
Anthropic bans AI model abuse, tightens rules on deception in updated Claude policy designed to target influence campaigns, surveillance and mistreatment of AI models https://www. databreachtoday.com/anthropic- bans-ai-model-abuse-tightens-rules-on-deception-a-33048
-
K
RE: https:// dair-community.social/@emilymb ender/117412868905734653 This thread! Read about Anthropic latest claims:
All 6 developments of Anthropic cuts Claude reasoning-token theft by up to 99%… →
NewswiresMastodonHacker NewsXRedditBlueskyGoogle News