US-Iran Talks Stalled, Likelihood of US Strike on Iran by End of March Rises to 69%

**Feb 28 – OpenAI Founder Sam Altman Announces DoD Partnership** OpenAI Founder Sam Altman announced on February 28 that the company is partnering with the U.S. Department of Defense (DoD) to deploy its AI models on a classified network. In all interactions, the DoD has demonstrated a strong focus on security and eagerness to collaborate for optimal outcomes. At the core of OpenAI’s mission is the **safe and broadly beneficial use of artificial intelligence**. Its two top security principles: 1. Prohibiting mass surveillance within the U.S.; 2. Holding individuals accountable for the use of force—including autonomous weapon systems. The DoD has acknowledged these principles and integrated them into its laws and policies, which are also included in the partnership agreement. OpenAI will build technical safeguards to ensure proper model functioning (a priority the DoD shares). It will deploy full-disk encryption (FDE) devices to support model operations and will only run

Feb 28 — Chain analytics provider Ai Yi (@ai_9684xtpa) reported that gold token XAUT recorded its largest single transaction in nearly three weeks earlier Tuesday. Four hours prior, the Tether Treasury address transferred 28,723 XAUT to UK asset manager Abraxas Capital Mgmt, valued at $151 million. Heka Funds (Abraxas Capital) is one of Tether’s largest and most critical institutional clients—previously holding 1.5% of total USDT issuance. Among Tether’s publicly disclosed on-chain addresses, it also ranks second for total interactive volume.

February 28, 2025 According to a report from the Hong Kong Commercial Daily, Hong Kong police recorded 12,505 online shopping fraud cases in 2025—an 8.2% year-over-year increase from 2024. The single largest fraud loss tied to online shopping last year involved cryptocurrency. A victim joined a Telegram crypto trading group and was messaged privately by a fake customer service rep, who demanded a “verification code” for group entry. The scammer then drained all virtual assets from the victim’s Telegram-linked wallet, leaving a total loss of over HK$3.7 million.

February 28: AI firm Anthropic issued a statement detailing two exceptions it has filed regarding the lawful use of its Claude AI model—banning large-scale domestic surveillance of U.S. citizens and fully autonomous weapon systems. To date, the company has not received direct communication from the U.S. Department of Defense (DoD) or the White House about the status of ongoing negotiations. The exceptions are rooted in two key considerations: 1. Current state-of-the-art AI models are not yet reliable for use in fully autonomous weapons, which would endanger U.S. military personnel and civilians; 2. Large-scale domestic surveillance violates fundamental human rights. Anthropic emphasized the DoD’s designation of the company as a supply chain risk is unprecedented, expressing deep disappointment. Since June 2024, the firm has continuously supported U.S. military personnel and remains committed to continuing this support. The company contends the designation is legally unf

Wallet 0x2942 was created 2 days ago and spent 50.75 $ETH($103K) to buy 6.7M $EDEL. Wallet 0x7a06 was created 7 hours ago and spent 30.72 $ETH($59K) to buy 3.17M $EDEL. Wallet 0xd365 was created 21 hours ago, received 100K $USDC to buy $EDEL, and has already spent 17.5K $USDC to buy 973,423 $EDEL. It still holds 82.5K $USDC and may keep buying.

On February 28th, Fortune reported that OpenAI founder Sam Altman stated the company is in discussions with the U.S. Department of Defense (DoD) about potential uses of its AI models and tools—no contract has been signed as yet. OpenAI will retain control over three key areas: implementation of its technology protection measures, the deployment targets for its models, and the regions where they will be deployed. It will also restrict deployment to cloud environments, not "edge systems"—in a military context, such systems may include aircraft and drones. As a significant concession, the U.S. government has agreed to incorporate OpenAI’s "red lines" into contract terms. These red lines prohibit AI use for autonomous weapon systems, domestic mass surveillance, and critical decision-making areas. Earlier today, Anthropic rejected the Pentagon’s request to relax security restrictions, prompting a Trump administration ban.