Portuguese 
English 
Spanish 
3 min of reading 
0 comments 
Anthropic advocates coordinated slowdown to allow more time for safety research and control of advanced AI models.
A proposal with significant impact in the technology sector has placed Anthropic, the company responsible for Claude, at the center of the global debate on artificial intelligence safety.
The company, based in San Francisco, United States, stated in a report released in June 2026 that a worldwide pause in AI advancement could help researchers, governments, and companies keep up with the pace of new systems.
According to Reuters and Associated Press, Anthropic assesses that increasingly powerful models could amplify risks if they advance faster than human oversight mechanisms.
-
Researchers develop “new cement” from pulp industry waste and concrete produced with açaĂ seeds and discarded tires, reducing environmental impacts without compromising the standards required by the construction industry.
-
5,300-year-old mummy surprises scientists after live fungi turn into artisanal bread: yeast extracted from Ătzi withstands time, digests aggressive chemical compound, and may reveal why the modern body lost part of its own ancestral microbiota
-
Error on the FIFA website releases free tickets for the 2026 World Cup for $0, surprising around 60 fans and now puts buyers against the clock to pay before cancellation.
-
Rains of up to 200 mm already have a date to advance over Brazil: atmospheric phenomenon, cyclone, and moisture river may cause intense storms in 6 states and put millions of people on alert in the coming days.
The company highlighted that a real pause would only be effective if major AI developers, especially in the United States and China, agreed to slow down simultaneously.
Anthropic report advocates coordinated brake on AI
The proposal was presented by Anthropic as an alternative to allow more time for artificial intelligence alignment research.
According to the company, it would be beneficial for the world to have the option to reduce or temporarily pause AI development.
This measure would allow social structures, governments, and experts to better keep pace with the rapid technology.
However, Anthropic itself acknowledged that an isolated pause could only weaken those who decided to slow down first.
In this scenario, competitors could advance faster and increase pressure on companies that prioritize safety.
United States and China enter the center of the dispute
Anthropic stated that a true pause would require an agreement among major AI companies in different countries.
This pact would need to include verifiable rules, capable of showing if everyone is complying with the slowdown.
The company cited the importance of coordination between the United States and China, two central hubs of the global technological race.
Without this international mechanism, companies and governments would continue facing difficult decisions between security, market, and geopolitics.
The race for AI advancement would continue to pressure decisions involving technical risks, commercial interests, and global strategy.
Proposal meets resistance in Washington and Silicon Valley
The idea faces strong resistance in the United States, especially among officials and technology executives.
American officials and leaders of major companies believe that slowing down AI could favor China in the technological race.
Industry representatives also argue that the fast pace maintains leadership, innovation, and competitive advantage.
Anthropic, on the other hand, argues that the lack of coordination can leave companies and governments facing risky choices.
The company tries to place artificial intelligence safety at the center of a discussion that mixes technology, market, and geopolitical power.
Trump authorizes preliminary evaluation of powerful models
President Donald Trump signed in June 2026 a decree related to the evaluation of advanced AI models.
The measure allows the American government to conduct preliminary analyses of the most powerful systems before launch.
According to Reuters, the goal is to observe risks in models developed by U.S. companies.
Anthropic’s proposal, however, goes beyond national oversight.
The company advocates for a global response, as cutting-edge artificial intelligence advances simultaneously in different countries.
Artificial intelligence safety becomes central point of the technological race
Anthropic’s proposal reinforces a growing dilemma in the technology sector.
On one side, companies compete for global leadership in increasingly advanced models.
On the other hand, experts advocate for more time to understand risks, limits, and forms of human control.
Anthropic itself admits that a global pause would be difficult, as it would depend on trust, oversight, and international cooperation.
Even so, the company states that the world should have this option in the face of the rapid advancement of AI.
After all, if artificial intelligence systems evolve faster than security structures, who will be able to ensure that they remain under human control?
Be the first to react!