AI This Week
Anthropic has employed the classic Game Boy game Pokémon Red to test its latest AI model, Claude 3.7 Sonnet. Unlike its predecessor, Claude 3.0 Sonnet, which struggled to leave the starting area, the updated model successfully battled three gym leaders, demonstrating impressive progress. Equipped with basic memory, screen pixel input, and function calls, Claude 3.7 Sonnet leveraged “extended thinking” to perform 35,000 actions and achieve significant milestones. The company revealed that within hours, the AI defeated Brock and subsequently conquered Misty, showcasing its advanced problem-solving capabilities. Pokémon Red joins a range of games now used to assess AI performance
Researchers from the University of Liverpool and the Research Institute for Diagnostic Accuracy, Netherlands, have demonstrated that artificial intelligence can significantly improve lung cancer screening efficiency. The study, published in the European Journal of Cancer, shows AI can accurately rule out negative low-dose CT scans, potentially reducing radiologists' workload by up to 79%. The AI tool, developed by Coreline Soft Co. Ltd., South Korea, successfully identified scans without significant lung nodules while ensuring all confirmed lung cancer cases were flagged for further review. This advancement could help optimize healthcare resources and ensure timely diagnoses in lung cancer screening programs worldwide.
New research reveals that 58% of businesses are adopting AI due to competitive pressure. The "Impact of Technology on the Workplace" report surveyed over 1,000 executives and found that most companies fear being left behind if they don't invest in AI. Other motivators include pressure from investors (41%) and stakeholders (30%). Currently, AI is mainly used for written tasks, data analysis, and customer support. While AI offers benefits like increased productivity, the report warns of potential downsides such as job losses as automation increases.
Together AI, backed by Nvidia, has raised $305 million in a funding round led by General Catalyst. This brings the artificial intelligence startup's valuation to $3.3 billion, more than doubling from last year. The funding will support expansion of Together AI's cloud platform and deployment of Nvidia's latest GPUs. The company's platform allows developers to train and deploy AI models, with over 450,000 users in 2024. Other investors include Salesforce Ventures and Coatue.
Alibaba plans to invest US$53 billion in AI infrastructure over the next three years, making it the largest private computing project of its kind. This investment, which exceeds Alibaba's total AI spending in the past decade, is more than half of the US's planned $100 billion Stargate project. The move underscores Alibaba's focus on AI-driven growth and its position as a leading global cloud provider. CEO Eddie Wu emphasized aggressive investment in AI and cloud computing infrastructure.
SnapLogic's recent report "AI Agents: The Final Frontier of the Enterprise" reveals that 84% of IT leaders trust AI agents equally or more than humans. The study, which surveyed over 1,000 IT decision-makers across four countries, shows high confidence in AI agents' ability to deliver business outcomes. Despite widespread interest, only 50% of organizations are currently using or planning to implement AI agents due to concerns about data security and integration challenges. The report highlights the potential of AI agents to transform enterprises while identifying key barriers to adoption.
AI adoption in broadcasting has accelerated, according to Haivision's 2025 Broadcast Transformation Report. The survey found 25% of broadcasters are now using AI, up from 9% in 2024. When asked about factors impacting the industry in the next five years, 64% named AI and machine learning. Top perceived AI benefits include efficiency gains, automated translation/captioning, and content creation. Among broadcasters not yet using AI, 41% plan to implement it within 2 years. The report is based on responses from about 900 broadcast and media professionals.
A new Nature study explores how artificial intelligence is transforming infectious disease modeling and prediction. Researchers found that AI-driven approaches can significantly improve outbreak forecasting and pathogen analysis. However, the study emphasizes that continued success depends on increased data transparency and reduced training costs. AI models have shown potential to accelerate epidemiological insights, support public health decision-making, and enhance genomic analysis of pathogens.
Over 20% of businesses are now spending between $50,000 and $100,000 annually on AI, with 9% investing over $500,000. It represents a rapid growth in AI adoption and expenditure compared to previous years, highlighting both the benefits and risks associated with this trend. Most businesses report satisfaction with their AI investments, citing high returns. However, concerns remain about cybersecurity and policy gaps.