OpenAI Confirms AI Safety Talks With Rivals as UK Lawmakers Seek Rules and Google Expands in Finland

Lehane acknowledged the three companies have been in talks for several weeks. Altman and others have noted the coordination could risk antitrust violations if it is found to suppress competition. Amodei proposed a narrow government waiver for such safety coordination, but Lehane said the firms do not need one, according to TechCrunch. The Information reported the three companies have been working to create an AI industry standards body, which Altman told staff would need to happen without U.S. government support. Hassabis called in July for a U.S. standards body with power to screen the most advanced models and coordinate industry-wide slowdowns if dangers progressed. President Trump has dismissed safety concerns as a hoax and said any slowdown would give China an advantage; his AI adviser David Sacks called fears of existential risk overblown. Lehane said OpenAI supports a provision in the FRONTIER Act that would require top frontier labs to allow independent verification organizations into their companies. TechCrunch said it reached out to Anthropic, Google and OpenAI for comment.

In the UK, Parliament's Joint Committee on Human Rights issued a report calling for an expansive AI bill with transparency and compliance enforced by an independent regulator, TechRadar reported. Committee chair Alex Sobel said no country, including the UK, has a current legislative and regulatory approach fit for purpose, and new legislation is needed across the AI supply chain and lifecycle. More than 70 MPs and peers have separately backed calls for Britain to prohibit the development and operation of AI, the report said, though artificial superintelligence remains hypothetical. OpenAI has argued for national AI safety regulation, including independent assessments and rules for slowing or stopping development, focused on the largest AI companies rather than every small firm. The UK's AI Security Institute already tests advanced models and has worked with frontier developers, but broader oversight still relies heavily on existing regulators and voluntary cooperation.

Google announced a record $15 billion investment in Finland, including three new data centers in Kajaani, Muhos and Vaala and an expansion of its Hamina facility, TechRadar reported. The project is expected to create around 37,000 jobs, with construction slated to begin in 2027. Under an agreement with Nordic energy company Fortum, up to 50% of output from the Loviisa nuclear power plant, which produces 10% of Finland's electricity, will be available to Google's data centers and AI infrastructure through 2050. Google said the investment is expected to add €3.6 billion a year to Finland's GDP and support Gemini, Google Maps, Search and YouTube. Finnish Prime Minister Petteri Orpo said the decision reflects Finland's strengths and that deepening collaboration with Google will deliver lasting benefits. Alphabet President and Chief Investment Officer Ruth Porat said Google is proud to deepen its roots in Finland and is pairing infrastructure expansion with new energy capacity, grid enhancements and energy affordability initiatives.

Google also said its technologies now support more than 300 languages spoken by 7 billion people, or 86% of the global population. In a blog post, it said Google Translate supports more than 250 languages, while Gemini 3.5 Live Translate powers real-time spoken translation across 70 languages and more than 2,000 language pairs. Gemini 3.5 Transcribe is its most precise speech-to-text model yet. Google's 1,000 Languages Initiative aims to support the world's 1,000 most-spoken languages, using its Universal Speech Model trained on 12 million hours of audio and cross-lingual transfer learning. The company cited open-data partnerships including WAXAL, covering 27 Sub-Saharan African languages; Project Vaani, with more than 30,000 hours of speech across 109 languages from over 155,000 speakers; and the Amplify Initiative, with more than 1,600 local experts and 20 universities. Its Language Explorer tool maps more than 7,000 spoken, written and signed languages.

On science, Google said AlphaFold has predicted all 200 million known protein structures and is used by 4 million researchers in 190 countries. AlphaGenome Atlas mapped all 9 billion possible single-letter genetic changes across the human genome. WeatherNext 3 delivers 50% more accurate precipitation forecasts a day or more ahead. Its Planetary Prediction Engine, combining global health, food security and socioeconomic data, has been used in the Ebola outbreak in the Democratic Republic of the Congo and in the U.S. to identify vulnerable communities across 21 CDC health indicators. Google also cited a breast cancer study with Imperial College London and the UK's NHS that found AI detected 25% of interval cancers previously missed in mammograms of 175,000 women, tuberculosis screening of more than 25,000 X-rays across 40 locations in six nations, and diabetic retinopathy screening support for more than 1.15 million cases, with plans to expand to 6 million over the next decade.

Editor's Summary

OpenAI, Anthropic and Google DeepMind have held weeks of AI safety talks as U.S. and U.K. lawmakers weigh new rules, while Google announced a $15 billion Finnish data-center investment and expanded language and science AI tools. The parallel moves show leading AI companies seeking to shape safety standards and infrastructure even as governments debate mandatory oversight.