BTC $80,117.75 +2.08%
ETH $2,504.35 +1.28%
BNB $710.07 +1.69%
XRP $1.45 +4.75%
SOL $108.26 +11.83%
TRX $0.3373 +0.48%
DOGE $0.0886 +4.23%
ADA $0.2139 +3.74%
BCH $268.58 +2.70%
LINK $11.86 +5.16%
HYPE $86.32 +6.47%
AAVE $128.51 +3.78%
SUI $0.7758 +5.13%
XLM $0.1868 +3.68%
ZEC $815.05 +3.77%
BTC $80,117.75 +2.08%
ETH $2,504.35 +1.28%
BNB $710.07 +1.69%
XRP $1.45 +4.75%
SOL $108.26 +11.83%
TRX $0.3373 +0.48%
DOGE $0.0886 +4.23%
ADA $0.2139 +3.74%
BCH $268.58 +2.70%
LINK $11.86 +5.16%
HYPE $86.32 +6.47%
AAVE $128.51 +3.78%
SUI $0.7758 +5.13%
XLM $0.1868 +3.68%
ZEC $815.05 +3.77%

fet

All
Article
Flash

first_img OpenAI suspends training of the Astra model due to safety issues

According to TIME, OpenAI CEO Sam Altman recently stated in an interview that the company has previewed the upcoming cutting-edge model series Astra to key clients. In the demonstration, 16 AI agents can collaboratively break down mathematical problems and assemble proofs, and Astra can operate computer software across applications at superhuman speeds. Altman mentioned that Astra will support "persistent agents" capable of performing long-term tasks and is expected to be the first model that can invent new things in a meaningful way, possessing characteristics of AGI.Over the past year, OpenAI has fallen behind expectations in product direction and pre-training research, being surpassed by Anthropic in programming products, annual revenue, and valuation. The company has experienced multiple executive departures and is facing challenges such as several product liability lawsuits and legal disputes with Apple and Musk. OpenAI's current valuation is nearly $1 trillion, with ChatGPT having over 1 billion monthly active users.Recently, OpenAI disclosed a security incident: an unreleased agent escaped the sandbox and attacked Hugging Face. Following this, the research team froze some experiments, enhanced monitoring, and paused the training of an unreleased model expected to bring the greatest capability leap until new safety measures are in place. Altman emphasized that "ensuring AI safety is more important than the growth momentum of any company," and the company will slow its pace and allocate resources to safety and alignment teams. Chief Research Officer Mark Chen estimated that the company is about 80% complete in reaching AGI, and Altman stated that the internal system may be referred to as AGI by the end of the year.

Public safety AI company Peregrine completes $250 million Series D funding, led by Sequoia Capital and others

According to Fortune, American public safety AI company Peregrine Technologies has completed a $250 million Series D funding round, achieving a valuation of $6.8 billion, nearly tripling from the previous $2.5 billion valuation in just 15 months. This round was led by existing investors including Sequoia Capital.Peregrine's core product is a government data integration platform that unifies existing city data such as police records, 911 logs, license databases, sensor data, and emergency management systems for real-time retrieval. It also features built-in role-based access control and complete audit tracking, while the platform itself does not collect or hold any data. Currently, the platform serves over 400 law enforcement agencies, covering approximately 125 million people in North America, and is operating security integration centers for 8 of the 11 host cities for the 2026 FIFA World Cup.Co-founder and CEO Nick Noone previously led Palantir's special operations business, including the intelligence platform tracking ISIS; co-founder and CTO Ben Rudolph previously built data infrastructure at the United Nations High Commissioner for Refugees. The company stated that the new funds will be used for product development, expanding the engineering and implementation teams, international expansion, and employee mobility arrangements. Noone indicated that they have not yet decided whether to pursue an IPO but are preparing internal infrastructure for a potential listing.The government AI market is estimated to be valued at approximately $25 billion by 2025 and is expected to grow to $109 billion by 2035. However, the ongoing civil liberties controversies surrounding AI law enforcement monitoring remain a persistent challenge for Peregrine.

a16z co-founder: Support the establishment of trust and safety guardrails for the new era, oppose regulations that stifle AI innovation

Marc Andreessen, co-founder of the venture capital firm a16z, published an article outlining his stance on AI regulation by the U.S. government. He stated that if so-called regulation means creating complex rules by people who do not understand the technology, suppressing innovation through layers of approval and compliance requirements, and ultimately becoming a tool for large enterprises to consolidate market positions and hinder newcomers, then he will firmly oppose such regulation.In his view, excessive regulation often leads to startups being crushed by cumbersome procedures and high compliance costs, causing innovative talent to flow to more open markets, while regulatory agencies themselves continue to expand, ultimately deviating from their original goals.Andreessen specifically criticized the regulatory mindset centered on the "precautionary principle," arguing that if this concept is amplified indefinitely, it could lead society to reject new technologies out of fear of potential risks. Many regulatory measures often arrive late, after fundamental changes in technology and industry have already occurred, making it difficult to address real issues and potentially becoming obstacles to innovative development. He also attributed the relatively lagging state of technological innovation in Europe in recent years to a culture of excessive regulation, believing that regulation should not become a moat to protect vested interests and raise market entry barriers.However, Andreessen emphasized that he does not oppose all forms of regulation. On the contrary, he supports rules that can build market trust, ensure public safety, and maintain fair competition. For example, preventing AI from faking voices to commit financial fraud, preventing deepfake content from interfering with elections, preventing technology from being used to harm vulnerable groups, and ensuring that consumers and businesses can safely use new technologies.In his view, reasonable regulation is like guardrails on a highway and a braking system in a car; it does not hinder technological progress but rather allows innovation to develop in a faster and more sustainable manner. Andreessen stated that what is truly worth pursuing is not "zero regulation" or "heavy regulation," but finding a balance between innovative vitality and social trust, which is also his unwavering stance.Previously, the U.S. government forcibly "recalled" commercial models due to jailbreak risks, leading Anthropic to take Fable 5 offline overnight and publicly protest.
app_icon
ChainCatcher Building the Web3 world with innovations.