BULLA plunged over 50% in the past 24 hours, and Binance will activate the Last Traded Price (LTP) protection mechanism for its USDT-margined perpetual contracts.

Self-custody platform SecondFi announced in a post that it has identified NIGHT balances held by some users at the time of the incident, which were recorded via snapshots in SecondFi or Yoroi wallets, were later emptied, while recoverable balances were not fully displayed. Some affected users received less than their entitled amount during asset recovery, and the difference will be remitted to their newly designated wallets. SecondFi noted that users who have already submitted recovery applications do not need to reapply, while those who have not yet done so may follow the original process, and recoverable balances will be corrected today. The official emphasized that this incident is unrelated to NIGHT airdrops and DeFi positions, users do not need to take additional actions, and warned against phishing links.

Financial Times analysis shows that since April 2026, the highest-earning account on Polymarket is a sports prediction account operating under the alias swisstony, which has generated approximately $12.25 million in profits. After analyzing 170 million predictions from around 1.2 million retail accounts, the FT found the list includes both high-stakes bettors who landed long-shot, high-odds predictions and high-frequency trading accounts. Swisstony has opened roughly 184,000 positions across more than 113,000 markets, with total investments of about $657 million. Since June, the account has made at least one prediction almost every hour for 26 consecutive days.

According to Bitget market data, Unitree Robotics’ gains further expanded to over 15%, with its share price crossing 730 yuan. Meanwhile, the A-share humanoid robot sector index rallied more than 5.5%, sparking a limit-up spree among constituent stocks. Hengfeng Tool, Ampleon, Fuhan Microelectronics, Weichuang Electric, Haohun Auto Electric, Henggong Precision, Bixin Microelectronics, and Danghong Technology all notched 20% daily limit-ups; Shangwei New Materials delivered a strong performance with two limit-up boards in three trading days.

Google parent company Alphabet is set to release its earnings after U.S. stock market close on Wednesday. The growth rate of its cloud business and the expansion of external TPU supply will be key points to watch, while its 2027 capital expenditure guidance and pressure on free cash flow turning negative will be the core focus of the report. For Q3 of fiscal 2026, single-quarter revenue is projected at $127.833 billion, with an expected year-on-year increase of 24.9%; net profit is forecast at $37.334 billion, up 6.73% year-on-year. Major investment banks have also released their outlooks: JPMorgan raised its year-on-year growth forecast for cloud business revenue from 88% to 100%; Deutsche Bank notes its Q3 search revenue growth will be relatively healthy; HSBC Research projects its operating margin to hit 44.5%, exceeding market expectations, with GAAP earnings before interest and taxes (EBIT) rising 35% year-on-year.

Binance founder CZ said four data breaches were reported today: Coinbase’s vendor Amplitude exposed some user contact data (funds were not affected); South Korean crypto exchange Bithumb suffered a user data leak. In traditional finance, two South Korean securities firms also experienced data leaks: Mirae Asset and Daishin Securities. While data breaches cannot be prevented, users can take steps to protect themselves. The leaked exchange data—including names, emails, phone numbers, addresses, ID documents, and transaction records—is mostly used for phishing, SIM swapping, and social engineering attacks. Users should limit the information they share, and use unique email aliases for each exchange platform: this way, if a leak occurs, it cannot be linked to their other accounts, and they can also identify which service was compromised.

Beating AI News, Anthropic engineering manager Felix Rieseberg announced this month his team has eliminated traditional coding interviews in hiring. Rieseberg, who previously worked on Claude.ai, Cowork, and Claude Code Desktop, noted that in the past, interviewers would ask candidates to write code on the spot (such as completing a React programming problem) to assess their coding ability. But Claude now performs these tasks far better, and he observed that the highest-performing engineers on his team are not always the best at such exams. He also found that some entry-level engineers, with the help of AI, are already delivering better results than senior staff. In the past, new hires with ideas often had to draft proposals, convince supervisors, and wait for design and development resources; now they can directly build product prototypes with AI, backed by real-world results. There’s an internal example at Anthropic: employees have long used Claude Code in their terminals, but two recently joined engineers, Anthony Morris and Alice Zhao, built a more user-friendly graphical interface for it. The pair completed the prototype in just one weekend and launched the first public version about eight days later. Rieseberg advised young engineers not to focus solely on learning programming frameworks or practicing interview questions. Instead, they should spend more time understanding company business, identifying real problems, and using AI to build solutions. He believes AI is narrowing the technical gap between new and senior employees, meaning experience and tenure no longer hold the same advantage as before. However, after ditching traditional coding interviews, he has not yet found a better evaluation method. Currently, he mainly reviews candidates’ resumes and work samples, with a particular focus on small projects that demonstrate problem-solving skills and product judgment.