Anthropic Says Its AI Models Can End Conversations With Users to Protect Themselves - 1

Photo by Adrian González on Unsplash

Anthropic Says Its AI Models Can End Conversations With Users to Protect Themselves

  • Written by Andrea Miliani Former Tech News Expert
  • Fact-Checked by Sarah Frazier Former Content Manager

Anthropic said on Friday that it has given its AI models, Claude Opus 4 and 4.1, the ability to end conversations with users. The startup explained that the new feature would be used in rare cases where it is necessary to prevent harm—directed toward the AI model.

In a rush? Here are the quick facts:

  • Anthropic allowed Claude Opus 4 and 4.1 the ability to end conversations with users to protect themselves.
  • The new feature will be used as a last resort only when users insist on engaging in harmful interactions.
  • The ability is part of Anthropic’s AI welfare program.

According to the article published by Anthropic, the company released this update as part of its AI welfare program, a new area in AI research that considers an AI system’s “interests” or well-being. It clarified that while the potential moral status of AI systems is “uncertain,” it’s researching ways to mitigate risks to its AI model’s welfare.

“We recently gave Claude Opus 4 and 4.1 the ability to end conversations in our consumer chat interfaces,” wrote the company. “This ability is intended for use in rare, extreme cases of persistently harmful or abusive user interactions.”

Anthropic explained that its model Claude Opus 4, the company’s most advanced model released with safety warnings , showed during tests preference for avoiding harm—such as the creation of sexual content involving children or information that could lead to acts of terror or violence.

In cases where users repeatedly requested Claude to engage in harmful conversations, the chatbot refused to comply and tried to redirect the discussion. Now, the chatbot can refuse to answer and block the chat so users cannot continue the conversation—except in cases of imminent risk.

The company clarified that the conversation-ending ability will be used only as a last resort—most users will not be affected by this update—and that users can start a new conversation on another chat immediately.

“We’re treating this feature as an ongoing experiment and will continue refining our approach,” wrote Anthropic. “If users encounter a surprising use of the conversation-ending ability, we encourage them to submit feedback by reacting to Claude’s message with Thumbs or using the dedicated ‘Give feedback’ button.”

The startup has been previously working on other projects related to AI welfare. Last year, Anthropic hired researcher Kyle Fish to study and protect the “interests” of AI models.

Google To Pay A $36 Million Fine In Australia Over Anticompetitive Deals - 2

Photo by Nathana Rebouças on Unsplash

Google To Pay A $36 Million Fine In Australia Over Anticompetitive Deals

  • Written by Andrea Miliani Former Tech News Expert
  • Fact-Checked by Sarah Frazier Former Content Manager

Google has agreed to pay a $35.8 million fine—55 million Australian dollars—in Australia on Monday after the country’s consumer watchdog found the tech giant had harmed competition through its deals with local telecommunications companies.

In a rush? Here are the quick facts:

  • Google has agreed to pay a $35.8 million fine over antitrust accusations in Australia.
  • The tech giant had agreements with Australia’s largest telecommunications companies, Telstra and Optus, to pre-install Google Search on Android devices.
  • The court has yet to determine whether the proposed fine and agreement between the parties are appropriate.

According to the official press release , Australia’s Competition and Consumer Commission (ACCC) accused Google of anti-competitive business practices, including deals with Telstra and Optus—Australia’s largest telecommunications and tech services providers—to pre-install Google Search on Android mobile phones.

“The understandings, which were in place between December 2019 and March 2021, required Telstra and Optus to only pre-install Google Search on Android phones they sold to consumers, and not other search engines,” explained the agency. “In return, Telstra and Optus received a share of the revenue Google generated from ads displayed to consumers when they used Google Search on their Android phones.”

Google acknowledged the anti-competitive practices and agreed to pay the fine. The company also committed to avoiding similar arrangements and to removing the mandatory pre-installation of the Google Search app on mobile devices.

According to Reuters , Google said that it’s pleased to resolve concerns and collaborate with the ACCC. The arrangements with the telecommunications companies have been in place since 2017.

“We are committed to providing Android device makers more flexibility to pre-load browsers and search apps, while preserving the offerings and features that help them innovate, compete with Apple, and keep costs low,” said a spokesperson from Google.

The court has yet to determine whether the proposed fine and agreement between the parties are appropriate.

Google is also facing antitrust challenges in other jurisdictions. In the European Union, regulators are investigating its AI Overview tool , while in the United Kingdom, the antitrust watchdog recently announced stricter measures to assess Google’s dominance in search services. Since 2024, the tech company has also been in a legal antitrust battle with the U.S. Department of Justice .