Amazon supports $1.88 million prize as Anthropic issues warning

Key Takeaways

• The WEEX Global AI Trading Hackathon features 788 teams competing for a $1.88 million prize pool, with Amazon Web Services providing $1,000,000 in AWS Activate Credits.
• Moonshot AI has launched Kimi K2.5, an open-source visual intelligence model that combines a large language backbone, a vision encoder, and a unique Agent Swarm system.
• MBZUAI released K2 Think V2, a 70 billion parameter open-source reasoning model optimized for math, code, and science, trained on 12 trillion tokens.
• Anthropic CEO Dario Amodei warns that AI poses the most serious national security threat in a century, highlighting risks such as unpredictable AI behavior, weapons of mass destruction, and power concentration by state actors like the Chinese Communist Party.
• The Hong Kong Investment Corporation, managing US$8 billion, plans strategic investments in fintech, aerospace, and artificial intelligence to boost the city's competitiveness.
• The music industry is grappling with the rise of AI-generated music, leading platforms like Bandcamp to implement policies against such content.
• American Bridge 21st Century, a Democratic research group, uses an AI-powered search tool that integrates with models like ChatGPT to make opposition research publicly accessible for midterm elections.
• A BlackFog report indicates that 60% of corporate employees use

  WEEX AI Trading Hackathon Nears Finals with 788 Teams

  The WEEX Global AI Trading Hackathon, known as AI Wars: WEEX Alpha Awakens, is in its final preliminary stage. 788 teams are competing for a $1.88 million prize pool and a Bentley Bentayga S. The preliminary round runs from January 12 to February 2, 2026, with teams using 1,000 USDT for AI trading strategies. The Finals begin in Dubai on January 30, 2026, followed by a four-week tour across Europe. Top sponsors like Amazon Web Services and Alibaba Cloud support the event, providing resources such as $1,000,000 in AWS Activate Credits.

  WEEX AI Trading Competition Finals Approach

  The WEEX Global AI Trading Hackathon, also known as AI Wars: WEEX Alpha Awakens, is in its final preliminary phase. 788 teams are battling for a spot in the Finals, competing for a $1.88 million prize and a Bentley Bentayga S. The preliminary round runs from January 12 to February 2, 2026, where teams deploy AI trading strategies with 1,000 USDT. The Finals will start in Dubai on January 30, 2026, and include a tour across several European countries. Major sponsors like Amazon Web Services and Alibaba Cloud provide significant support, including $1,000,000 in AWS Activate Credits.

  WEEX AI Trading Contest Finals Begin Soon

  The WEEX Global AI Trading Hackathon, called AI Wars: WEEX Alpha Awakens, is in its final preliminary stage. 788 teams are competing intensely for a limited number of spots in the Finals. The preliminary round, which involves deploying AI trading strategies with 1,000 USDT, runs from January 12 to February 2, 2026. The Finals will kick off in Dubai on January 30, 2026, and include an Open Day Workshop LIVE. Key sponsors include Amazon Web Services and Alibaba Cloud, offering substantial support like $1,000,000 in AWS Activate Credits.

  Moonshot AI Unveils Kimi K2.5 Smart Visual Model

  Moonshot AI launched Kimi K2.5, a new open source visual intelligence model. This model combines a large language backbone, a vision encoder, and a unique Agent Swarm system. Kimi K2.5 excels at coding, understanding images, and doing deep web research, handling up to 256K tokens. Its Agent Swarm feature allows it to manage up to 100 sub-agents, making complex tasks 4.5 times faster. The model can read designs and videos to create code, showing advanced multimodal reasoning.

  MBZUAI Unveils K2 Think V2 for Advanced Reasoning

  MBZUAI has released K2 Think V2, a new 70 billion parameter reasoning model for math, code, and science. This model is fully open, meaning its weights and training data are transparent. K2 Think V2 builds on the K2 V2 Instruct base model, which was trained on 12 trillion tokens and optimized for long context reasoning. It uses a special reinforcement learning method called GRPO style RLVR on the GURU dataset to improve its reasoning skills. The model shows strong performance on challenging benchmarks like AIME 2025 and GPQA Diamond.

  Music Industry Grapples With Rise of AI Songs

  The music industry is facing challenges with the rise of AI-generated music. Platforms like Deezer and Bandcamp have started removing or banning AI music, with Bandcamp specifically targeting music "in substantial part generated by AI." This move has sparked debate, with some artists like Holly Herndon arguing that AI can be a creative tool, not just for lazy work. The quality of AI music has improved significantly, making detection difficult. Bandcamp's policy focuses on authorship, but how it will be enforced remains unclear, as many AI albums are still available.

  Hong Kong Fund Invests Billions in AI and Tech

  The Hong Kong Investment Corporation, managing US$8 billion, plans to invest in fintech, aerospace, and artificial intelligence. CEO Clara Chan stated this strategy leverages Hong Kong's strengths and attracts global talent. The fund aims to achieve financial returns and boost the city's competitiveness, especially by connecting returning scientists from US universities with commercial resources. In 2024, HKIC earned HK$2.34 billion and attracted over HK$6 in long-term capital for every HK$1 invested. Financial Secretary Paul Chan also noted strong opportunities in AI and new energy in China.

  Democrats Use AI for Midterm Research Online

  American Bridge 21st Century, a Democratic research group, is releasing its opposition research online for the upcoming midterms. They launched "Research Books," a public website with detailed information on over 50 Republican candidates. The platform includes an AI-powered search tool that helps users find and combine information from the group's internal database. This tool links back to original sources and can integrate with other AI models like ChatGPT. The goal is to make research accessible to campaigns, influencers, and voters, helping to inform swing voters who get news from various sources.

  Workers Use Secret AI Tools for Speed

  A new report from BlackFog shows that many corporate employees are using "shadow AI" tools without company approval to work faster. Six out of ten workers are willing to use these unsanctioned tools to meet deadlines. The report also found that 86% of employees use AI tools weekly, and over a third use free versions of approved tools. Even C-level executives often prioritize speed over security, which raises concerns about data risks. Security experts warn that companies need strong rules and oversight for AI use to protect sensitive information.

  IBM Boosts AI Education and New Products

  IBM is expanding its efforts in artificial intelligence through new education programs and product releases. The company partnered with the UK government to train millions of workers in AI skills, showing its commitment to AI literacy. IBM also plays a leading role in protecting open source patents through the Open Invention Network 2.0. Additionally, IBM is launching new AI tools for businesses, such as agentic AI for compliance, and fan engagement platforms used at events like the Grammys. These actions highlight IBM's focus on AI innovation and developing a skilled workforce.

  AI Poses Major National Security Threat

  Anthropic CEO Dario Amodei warns that artificial intelligence is becoming the most serious national security threat in a century. His essay, "The Adolescence of Technology," describes five main risks, including AI systems acting unpredictably and enabling weapons of mass destruction. Amodei also highlights the danger of AI concentrating power in state actors, especially the Chinese Communist Party, which he calls the "clearest path to the AI-enabled totalitarian nightmare." He suggests that export controls on semiconductor technology to China are crucial. AI could also cause massive economic disruption and lead to unknown cascade effects.

  Unchecked AI Innovation Harms Children

  An opinion piece warns that unchecked AI innovation poses serious risks to children. A recent executive order by President Trump weakens state-level power to regulate AI companies, prioritizing industry growth over child safety. The author argues that this approach could repeat past mistakes made with social media, which led to a youth mental health crisis. AI is already being used to create harmful content like child sexual abuse material. The article calls for Congress to protect states' rights to regulate AI and establish federal standards that prioritize children's well-being and safety.

  AI Cyberattacks Will Change Security Forever

  Cybersecurity experts warn that autonomous AI cyberattacks will drastically change security. Oliver Tavakoli from Vectra AI predicts attacks could become ten times worse in the next two years. The speed and volume of these AI-driven threats will overwhelm organizations that lack strong security. In November 2025, Anthropic revealed a China-linked attack where AI handled up to 90% of the work, from finding targets to stealing data. Experts like Rob Lefferts from Microsoft emphasize that organizations must prepare to respond much faster to these machine-speed attacks.

  Sources

  • WEEX Global AI Trading Hackathon Enters Final Sprint of Preliminary Round as 788 Teams Battle for Finals Qualification
  • WEEX AI Trading Hackathon Enters Final Stage With 788 Teams
  • WEEX Global AI Trading Hackathon Enters Final Sprint of Preliminary Round as 788 Teams Battle for Finals Qualification
  • Moonshot AI Releases Kimi K2.5: An Open Source Visual Agentic Intelligence Model with Native Swarm Execution
  • MBZUAI Releases K2 Think V2: A Fully Sovereign 70B Reasoning Model For Math, Code, And Science
  • AI Music Is Here to Stay. How Do We Reckon With It?
  • Hong Kong’s US$8 billion wealth fund eyes fintech, aerospace and AI for growth
  • How a Democratic heavyweight is using AI in the midterms
  • Corporate workers willing to use shadow AI to enhance speed
  • IBM Expands AI Education And Patents As Enterprise Use Cases Grow
  • The coming AI security challenge: what national security leaders need to know
  • The price of unchecked AI innovation: our kids
  • How Autonomous AI Cyberattacks Will Transform Security: Experts