According to a Bloomberg report on September 10, a Nasdaq-affiliated investment arm will invest $100 million in Payward, the parent company of cryptocurrency trading platform Kraken, bringing the latter's valuation to $21 billion.

This investment continues the partnership reached between the two parties in March of this year. As part of the partnership, Kraken will offer users tokenized shares of Nasdaq-listed companies, granting users the same voting rights as ordinary shareholders upon holding the relevant tokens.

Nasdaq is accelerating its efforts to tokenize its stocks and plans to launch its own stock token in the second quarter of 2027. The company applied to the U.S. Securities and Exchange Commission (SEC) last year to amend relevant rules to allow tokenized trading of its stocks on regulated exchanges, and received approval in March of this year.

This investment also demonstrates that the integration of traditional Wall Street financial infrastructure with the crypto market is accelerating.

According to Beating AI News, tech industry user leo posted on X that, according to his understanding, OpenAI is very close to verifying the Hodge Conjecture, and OpenAI or Anthropic may also be close to solving the Birch-Swinnerton-Dyer Conjecture.

The user emphasized that the news is currently just a rumor and its authenticity cannot be confirmed. However, if it is true, it means that three "Millennium Prize Problems" may be solved within a month. Currently, the Hodge Conjecture and the Bech and Swinnerton-Dale Conjecture are still unsolved Millennium Prize Problems.

Previous breakthroughs in mathematics by AI have already attracted market attention, and the prediction market currently lists the Hodge conjecture and the Birch and Swinnerton-Dale conjecture as the next millennium problems that AI may solve.

According to Beating AI News, OpenAI is pushing for nationwide mandatory AI safety rules in the United States, requiring Congress to directly regulate cutting-edge AI companies. The rules would include standardized testing, independent safety assessments, cybersecurity, and reporting of major incidents.

OpenAI hopes governments will establish unified safety thresholds, clearly defining when AI development must slow down or even stop, and promote the adoption of similar standards by various countries. This regulatory framework primarily targets a small number of large companies developing the most advanced models, excluding ordinary startups and small developers.

OpenAI also announced its support for four California AI safety laws, some of which it had not previously endorsed. OpenAI stated that it has changed its stance as its model capabilities have rapidly improved.