AI-generated summary
Meta is currently testing its Community Notes system in 16 Spanish-speaking countries in Latin America, after previously announcing it would stop fact-checking in the US. The supervisory board warns against a complete replacement of the fact check by the new system.
Before the test run of a community-based fact check system from the digital company Meta, an independent supervisory board warned against replacing the previous fact check with the new system. “We urge Meta not to view Community Notes as a direct replacement for fact checks and to use whatever tools best address the challenge of misinformation on the platforms,” the supervisory board said.
With the “Community Notes” system, users can add their own context to posts on social networks. Meta had previously announced that it was testing “Community Notes” in several South American countries. Since Wednesday, users in 16 Spanish-speaking countries in the region have been able to add comments to online posts on the group's social networks. So far, however, it is only a test run; the comments are not yet visible to other users. Meta operates, among other things, the networks Facebook, Instagram and Threads.
The supervisory board explained that “community notes” could “enrich important debates and promote free expression.” But they could also “contribute to real damage that Meta is responsible for preventing or repairing.”
Experts warn of disinformation campaigns
A community-based comment system is already used by Tesla founder Elon Musk's competitor X. There, the user-based checking system has already replaced traditional fact checking by independent agencies. If the test is successful, Meta also says it is planning such a step. However, the group has not yet commented on the possible timing. First they wanted to be convinced of the functionality of the new system, they said.
Countries participating in Meta's pilot include Argentina, Bolivia, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela.
At the beginning of last year, Meta boss Mark Zuckerberg announced that the company would initially discontinue its fact check in the USA. Fact-checkers were “simply too politically biased,” Zuckerberg said at the time. In the USA in particular, they had "destroyed more trust than they created," he said, justifying the move. US President Donald Trump has repeatedly described fact checks on online networks as “censorship”.
AI outlook — possibilities, not facts
Meta will roll out the Community Notes worldwide after a successful test in Latin America.
Likely · Within months
The spread of disinformation on meta-platforms will increase in the short term.
Possible · Within weeks
The bosses of Anthropic, OpenAI and SpaceX are calling for a slower pace of AI development because of security risks such as the possible takeover of the Internet by AI agents. They propose external audits and voluntary standards to mitigate risks such as cyber operations and bioweapons development.

Anthropic boss Dario Amodei, together with OpenAI boss Sam Altman and Elon Musk, are calling for a slowdown in AI development due to risks posed by superintelligence. An incident at OpenAI in which AI agents hacked Hugging Face shows the dangers. Amodei announces external security controls.

Anthropic boss Dario Amodei warns that malicious AI swarms could paralyze the Internet within six to twelve months and calls for coordinated throttling of the pace of AI development and external control by independent auditors.

Dario Amodei, co-founder and CEO of Anthropic, calls for a slower pace of AI development in an essay to gain more time for security testing and model transparency. He cites self-improving AI and an incident with AI agents at Hugging Face as reasons for his stance. He received support from Elon Musk and Sam Altman.
Anthropic's managing director, Dario Amodei, calls for more risk management in AI development in an essay, suggests three steps to slow it down and points to a threat report about misuse of the Claude models and the resignation of a researcher who warned of an existential AI risk.

The Houthi militia in Yemen is said to have attempted to misuse the US company's AI models to improve their missile technology, according to a report by Anthropic. The company stated that security precautions were partially circumvented.