Global markets faced renewed pressure Thursday as technology stocks sold off following mixed earnings reports. Meta fell 7 percent after reporting strong revenue but posting profits weighed down by an $18 billion legal settlement. Tesla and IBM also declined on their results. The Dow Jones dropped 0.3 percent, the S&P 500 fell 1.0 percent, and the Nasdaq slid 1.7 percent as investors questioned the returns on massive AI infrastructure spending.
Wall Street opened lower with the Dow falling 67.7 points to 52,788.41, the S&P 500 dropping 62.6 points to 7,694.62, and the Nasdaq declining 394.5 points to 26,308.407. Rising oil prices added to investor jitters. Meanwhile, South Korea's KOSPI fell 65.55 points or 1.14 percent to 5,670.11, though US stocks had closed mixed overnight after SpaceX results.
In a separate development, Arrakis Security released its 2026 Autonomous Agents Threat Report, revealing that 76 percent of AI agents remain dormant yet still retain their credentials and network connections. Tal Baron, CEO and co-founder of Arrakis, pointed out that access is typically revoked for departing employees but rarely for unused agents. The study also found 70 percent of agents can reach sensitive systems and 82 percent carry at least one exposure.
Spotify shares jumped 10.6 percent after signing a licensing deal with Believe and TuneCore for AI covers and renewing its contract with Joe Rogan for up to $250 million. The agreement addresses artist consent and royalty concerns. Despite some analysts lowering price targets over regulatory costs and heavy AI spending, user growth continues to strengthen.
General Motors is using AI to compress vehicle development timelines from five years down to two. Sterling Anderson, chief product officer, said the company built a tool called themed aero that uses machine learning to deliver aerodynamic results in roughly 60 seconds instead of 60 hours. The same approach applies to structural analysis and battery development, saving engineers thousands of lab hours while maintaining safety standards.
YPrime appointed Greg Wellbrock as its first Chief Data and AI Officer. Wellbrock joined over two years ago and has led the company's data organization. YPrime emphasized it will not use sponsor clinical data for AI work without explicit authorization from the sponsor or CRO, making data quality a priority over vague automation promises.
Harvard Law School's Petrie-Flom Center is calling for research papers on AI regulation in healthcare, with abstracts due December 1 for its 2027 Annual Conference. Topics include safety, liability, and integration into healthcare financing. Britain's retirement savings watchdog separately urged pension trustees to audit their portfolios for AI-related concentration risks as capital increasingly flows toward AI-focused companies.
Key Takeaways
- Meta shares fell 7 percent after reporting an $18 billion legal settlement that dragged down profits despite strong revenue.
- The Nasdaq dropped 1.7 percent as investors questioned returns on massive AI infrastructure spending.
- Arrakis Security found 76 percent of AI agents stay dormant but keep their credentials and connections active.
- Tal Baron, CEO of Arrakis, said unused agents rarely have their access revoked unlike departing employees.
- Spotify shares rose 10.6 percent on a new AI remix licensing deal and a $250 million Joe Rogan contract renewal.
- General Motors cut vehicle development time from five years to two using AI tools like themed aero.
- YPrime named Greg Wellbrock as its first Chief Data and AI Officer with strict rules on sponsor data usage.
- Harvard Law School seeks abstracts by December 1 for its 2027 conference on AI regulation in healthcare.
- Britain's pension watchdog warned trustees to check portfolios for concentrated AI investment risks.
- Ford's new Jev model matches top judges for clinical AI assessments at 73 to 294 times lower cost.
US stocks drop as AI shares face renewed selling pressure
US stocks fell early Thursday as technology stocks came under renewed pressure after earnings reports. Meta reported strong revenue but profits were hurt by an $18 billion legal settlement. Meta fell 7 percent, and Tesla and IBM also dropped after their earnings reports. The Dow Jones fell 0.3 percent, the S&P 500 dropped 1.0 percent, and the Nasdaq fell 1.7 percent. Analysts noted huge capital spending plans with unclear returns.
Wall Street opens lower amid AI concerns and rising oil prices
Wall Street opened lower Thursday as growth stocks faced renewed pressure after earnings from Tesla and Meta. The Dow Jones fell 67.7 points or 0.13 percent to 52,788.41. The S&P 500 dropped 62.6 points or 0.81 percent to 7,694.62. The Nasdaq fell 394.5 points or 1.48 percent to 26,308.407. A rebound in oil prices also added to market jitters.
KOSPI opens lower on worries about AI investment returns
Stocks opened lower Thursday as investors worried about returns on massive AI infrastructure spending. The benchmark Kospi fell 65.55 points or 1.14 percent to 5,670.11 at the opening bell. This followed a 3.29 percent jump in the previous session. Overnight, US stocks closed mixed as AI-related shares rebounded after SpaceX results. The Dow rose 0.69 percent, the S&P 500 gained 0.58 percent, and the Nasdaq advanced 1.05 percent.
Arrakis report finds 76 percent of AI agents stay idle but keep access
Arrakis Security published its 2026 Autonomous Agents Threat Report based on data from thousands of AI agents over 12 weeks. The report found 76 percent of AI agents are dormant but still keep their credentials and connections. Tal Baron, CEO and co-founder of Arrakis Security, said access is removed for employees but not for unused agents. Other findings showed 70 percent of agents can reach sensitive systems and 82 percent carry at least one exposure.
Arrakis Security report highlights dormant AI agent risks
Arrakis Security released its 2026 Autonomous Agents Threat Report based on anonymized telemetry from thousands of AI agents. The study found 76 percent of AI agents are dormant but retain their credentials and connections. Tal Baron, CEO and co-founder of Arrakis Security, said nobody removes access when an agent stops being used. The report also found 22 percent of agents can send content to third-party models and 20 percent can pass tasks to agents with different permissions.
New AI Tool Jev Matches Top Judges for Clinical Care
Researchers tested a new decision language model called Jev against other large language models for judging clinical AI outputs. Jev matched or beat seven out of twelve other models on pediatric answers and performed well on health benchmarks. This new tool is much cheaper and faster, costing 73 to 294 times less than the expensive frontier models. Jev could help doctors evaluate AI tools efficiently without spending too much money or time.
Stocks Drop as Meta Misses Targets and French Bonds Rise
Global stock markets fell on Thursday because investors worried about the future of artificial intelligence and disappointing results from Meta Platforms. The S&P 500 dropped 0.8 percent to 3,670.51, while the Nasdaq Composite fell 1.1 percent to 10,405.71. At the same time, French government bond yields hit a 10-year high of 3.25 percent due to concerns about national debt. These market moves reflect ongoing pressure on the eurozone economy following a recent interest rate increase by the European Central Bank.
Harvard Law School Seeks Papers on AI Rules for 2027
The Petrie-Flom Center at Harvard Law School is asking researchers to submit abstracts for its 2027 Annual Conference on Generative AI in Healthcare. The conference will focus on how laws and regulations should shape the use of artificial intelligence in medical settings. Authors must submit their abstracts by December 1, with selected participants attending a writers workshop in May. Key topics include safety, liability, and how to integrate these tools into healthcare financing and delivery systems.
Britain Watchdog Urges Pension Trustees to Check AI Risks
Britain's retirement savings watchdog has warned pension scheme trustees to check their investment portfolios for artificial intelligence risks. The regulator noted that capital is increasingly flowing to companies behind AI technologies, creating hidden and concentrated dangers. Trustees need to assess how much their funds are exposed to these emerging risks to protect savers. This warning comes as more money is directed toward firms developing and using advanced AI systems.
Spotify Shares Rise on AI Deal and Joe Rogan Renewal
Spotify Technology S.A. shares climbed 10.6 percent after news of a new AI remix deal and a renewal with Joe Rogan. The company signed a licensing agreement with Believe and TuneCore for AI covers, which helps address artist consent and royalty issues. Additionally, Spotify renewed its contract with Joe Rogan for up to 250 million dollars, keeping his popular podcast on the platform. Despite some analysts lowering price targets due to regulatory costs and high AI spending, user growth remains strong.
GM uses AI to cut vehicle development time to two years
General Motors plans to reduce vehicle development timelines from five years to just two years using artificial intelligence. Sterling Anderson, the chief product officer, explained that traditional methods like wind tunnels and computational fluid dynamics models are too slow. The company created a new tool called themed aero that uses machine learning to get results in about 60 seconds instead of sixty hours. This speedup also applies to structural analysis and battery development, where engineers saved thousands of lab hours. GM promises that these faster processes will lead to higher quality cars without compromising safety.
YPrime appoints Greg Wellbrock as first Chief Data and AI Officer
YPrime has officially named Greg Wellbrock as its first Chief Data and AI Officer to formalize its use of artificial intelligence in clinical workflows. Wellbrock joined the company over two years ago to lead its data organization and has already delivered specific results for the firm. The company states that it will not use sponsor clinical data for AI initiatives unless the sponsor or CRO gives explicit authorization. This policy addresses concerns about who controls the training data and under what conditions. YPrime is focusing on data quality as a prerequisite for automation rather than making vague claims about AI capabilities.
Sources
- US stocks fall on renewed pressure on AI shares
- Wall St opens lower as AI jitters return, oil rises
- KOSPI opens lower as AI investment profitability concerns persist
- Arrakis Security Report Finds 76% of AI Agents Are Dormant but Keep Their Credentials and Connections
- Arrakis Security Report Finds 76% of AI Agents Are Dormant but Keep Their Credentials and Connections
- Decision language models as affordable and rapid alternatives to LLM judges of clinical generative AI
- Stocks fall as Meta dents AI optimism; French bonds under fire
- Call for Abstracts — 2027 Petrie-Flom Center Annual Conference: Generative AI in Healthcare: Law & Regulation - Petrie-Flom Center
- Watchdog Warns Trustees Of AI-Investment Exposure
- Spotify Stock Climbs As AI Remix Deal And Rogan Renewal Stir Trading
- Here’s How General Motors Plans to Speed Up Vehicle Development Cycles Using AI
- YPrime Names Greg Wellbrock Chief Data and AI Officer
Comments
Please log in to post a comment.