Nvidia Falls Up to 7.4% as Apple Studies AI Health Questions

AI-linked shares feature prominently in several reports on market declines, though the figures vary by trading day and article. Nvidia falls between 4.4% and 7.4% in accounts of sharp Monday selloffs, while Alphabet’s Google and Microsoft also lose ground in some sessions. One report puts the S&P 500 down 2%, the Nasdaq down 3.4% and the Dow down 1.7%, with all 11 S&P 500 sectors lower as investors weigh U.S.-China trade tensions. Other accounts cite recession fears, inflation and interest rates; a separate report describes Nvidia and CoreWeave falling in premarket trading amid questions about AI’s economic effects. Across the coverage, Nvidia’s weaker-than-expected earnings and concerns over high valuations and competition also feature. Apple and Siri are not involved in these market reports.

Apple is studying how people use AI to answer health questions through a research app developed with healthcare AI company OpenEvidence. The study is open to eligible U.S. users over 21 with an iPhone 15 or later, iOS 26 or later, and Apple Intelligence enabled; Apple aims to enroll as many as 200,000 participants for more than 12 months. Users can ask questions, rate simplified answers based on clinical sources, complete surveys and optionally share HealthKit data. Apple says participant data will be anonymized and not used to train AI models, and OpenEvidence will delete question-and-answer data after 24 hours. Apple says the app is separate from Siri and Apple Intelligence; the study is led by Health and Fitness vice president Sumbul Desai.

Several articles focus on practical uses and governance for AI. A mapping guide explains how PMTiles and HTTP Range requests can serve map data from a single file: it estimates about $0.05 per 1,000 city-zoom map loads on Bunny.net, compared with $5 per 1,000 Mapbox loads after its free tier. A marketing article recommends keeping people responsible for strategy and creative decisions while using AI for research, brief preparation and approved asset variations. Akeneo warns that retailers adding AI agents need reliable product data, shared governance, human review and clear change records to avoid disconnected systems. Thomson Reuters CEO Steve Hasker calls AI an opportunity and says the company will continue investing in AI-driven innovation.

Research and product updates include IBMARL, a multi-agent learning method combining demonstrations and reinforcement learning. Across four cooperative benchmarks, it beats imitation policies 2.6 to 4.2 times faster than demonstration-based baselines on two tasks, is the only method to beat them on two others, and surpasses the demonstrator on two tasks. GitHub also puts bulk AI Scan in public preview, letting teams launch security scans on up to 20 repositories at once. Scans check default branches, including languages CodeQL does not support, and results join the code-scanning backlog; a Copilot license is not required.

Jamaica’s government approves a memorandum of understanding with investors to explore AI infrastructure investment. Prime Minister Andrew Holness says the aim is to help the country become an AI technology developer and create future digital-economy jobs, with advice from the government’s AI Task Force. The investors, proposed investment amount and timeline have not been disclosed.

Key Takeaways

  • Reports describe Nvidia falling 4.4%–7.4% in Monday selloffs, with Google and Microsoft also declining in some accounts; recession, rates, inflation and U.S.-China trade concerns weigh on markets.
  • One reported session sees the S&P 500 fall 2%, the Nasdaq 3.4% and the Dow 1.7%, with all 11 S&P 500 sectors lower.
  • Apple and OpenEvidence are studying AI health-question use among eligible U.S. iPhone users, with a target of up to 200,000 participants over more than 12 months.
  • Apple says study data will not train AI models; OpenEvidence deletes question-and-answer data after 24 hours. Apple says the app is separate from Siri and Apple Intelligence.
  • A PMTiles mapping guide estimates Bunny.net delivery at about $0.05 per 1,000 city-zoom map loads, versus $5 per 1,000 Mapbox loads after its free tier.
  • IBMARL combines demonstrations and reinforcement learning; across four cooperative benchmarks it outperforms the demonstrator on two tasks.
  • GitHub’s public-preview AI Scan bulk launch covers up to 20 repositories and does not require a Copilot license.
  • Akeneo says retailers using AI agents need trusted product data, shared governance, human review and change records.
  • Jamaica approves an MOU to explore AI infrastructure investment, but the investors, funding amount and timeline remain undisclosed.
  • Thomson Reuters CEO Steve Hasker describes AI as an opportunity and says the company will keep investing in AI-driven innovation.

Nvidia and AI stock losses pull US markets lower

The S&P 500 fell 0.7% on Monday, March 27, 2023, and faced a third straight loss. Nvidia fell 5.4%, while Alphabet’s Google and Microsoft also declined. Investors worried about a possible recession and how rising interest rates could affect the economy. The S&P 500 was down 10% for the year, and the Nasdaq composite had fallen 12%.

Nvidia and other AI stocks extend market losses

The S&P 500 fell 0.7% Monday, with Nvidia dropping 4.4% and leading the decline. Palantir fell 3.2% and C3.ai dropped 2.8%. Investors worried about rising interest rates and their effect on technology stocks. The article also reports recent losses across the S&P 500 and Nasdaq, citing interest rates, inflation concerns and a falling dollar as factors.

Nvidia-led losses weigh on US stocks and other markets

The S&P 500 fell 0.7% Monday, led by Nvidia, which dropped 4.4% after reporting weaker-than-expected earnings. Alphabet and Microsoft shares also fell as investors worried about a possible recession. The Dow Jones industrial average lost 0.6%, the Nasdaq composite fell 0.8%, and the Russell 2000 dropped 1.1%. The 10-year Treasury yield fell to 2.6%, while the dollar rose against the yen, gold fell to $1,646.90 an ounce and oil slipped to $62.45 a barrel. Technology stocks led longer-term losses, with the sector down 14.1% over the past year.

Nvidia and AI stock declines drag major US indexes lower

The S&P 500 fell 0.9% Monday, while the Dow Jones Industrial Average dropped 0.9% and the Nasdaq composite lost 1.1%. Nvidia fell 5.4% after reporting weaker-than-expected earnings. Alphabet’s Google dropped 2.1% and Microsoft fell 1.5%. Investors worried about a possible recession, rising Treasury yields and trade tensions between the United States and China. The losses reversed the S&P 500’s 0.6% gain from the previous day.

Nvidia and AI stock losses pull the S&P 500 down 2%

The article’s headline says declines in Nvidia and other AI stocks pulled the S&P 500 down 2%. The supplied article content contains only a publication notice and copyright information, with no further details about the market decline.

Nvidia and AI stock declines push the S&P 500 down 2%

The S&P 500 fell 2% Monday and was on track for a third straight loss as investors worried about a possible recession. Nvidia dropped 6.4% to $65.03, while Alphabet's Google fell 3.4% to $26.19 and Microsoft lost 2.4% to $238.19. About 90% of S&P 500 stocks declined, though the technology sector rose 1.1% Monday and was up 5.6% so far this year. The S&P 500 has fallen 10% this year, and the Dow Jones industrial average has dropped 12%.

Nvidia and AI stock losses send the S&P 500 lower

The S&P 500 fell 1.9% Monday and was on track for a third straight loss as more declines in AI stocks, led by Nvidia, pulled the U.S. market lower. The article does not provide further details about the day's market moves.

Nvidia and AI stocks fall as trade tensions weigh on markets

The S&P 500 fell 2% Monday, its third straight loss, while the Nasdaq composite dropped 3.4% and the Dow Jones industrial average fell 1.7%. Nvidia fell 7.1% to $65.99, and Alphabet's Google declined 4.4% to $114.99. Investors worried about how ongoing trade tensions between the U.S. and China could affect the global economy. All 11 S&P 500 sectors fell, including technology, down 4.1%, communication services, down 3.4%, and consumer discretionary, down 3.3%.

Nvidia and AI stock declines lead broad market losses

The S&P 500 fell 2% Monday as Nvidia and other AI stocks dragged the U.S. market lower. The Nasdaq composite dropped 3.4% and the Dow Jones industrial average lost 1.7%. Nvidia, a major maker of computer chips for artificial intelligence and other uses, fell 7.1% to $65.99, while Alphabet's Google dropped 4.4% to $114.99. Investors were concerned about the effects of ongoing U.S. and China trade tensions on the global economy, and all 11 S&P 500 sectors declined.

Nvidia and AI stock declines pull the S&P 500 down 2%

The S&P 500 fell 2% Monday and was on track for a third straight loss, while the Nasdaq composite dropped 3.8%. Nvidia shares fell 7.4%, and the article also reports declines for Alphabet's DeepMind and Microsoft's Nuance Communications. It says high AI stock valuations and rising competition from other technologies contributed to the losses. The S&P 500 had fallen 10% over the past month, and the article said it was expected to fall further in the coming weeks.

Apple studies how people use AI for health questions

Apple is testing a research app that lets people ask health questions and rate the answers. The app gives simple responses with links to peer reviewed medical sources, and Apple hopes to use the study to improve easy to understand health tools. Apple says the app is not connected to Siri or Apple Intelligence, and participant data will stay private and will not train AI models. The study is led by Apple vice president of Health and Fitness Sumbul Desai, M.D.

Apple launches study app to track how people use AI for health

Apple and healthcare AI company OpenEvidence are launching a study app for U.S. users over 21 who have an iPhone 15 or later, iOS 26 or later, and Apple Intelligence enabled. Apple aims to enroll up to 200,000 people for more than 12 months. Users can ask health questions and get simplified answers based on clinical sources such as The New England Journal of Medicine, the Journal of the American Medical Association, and Nature. Participants can rate answers, complete monthly surveys, and share HealthKit data for more tailored suggestions; Apple says data will be anonymized, will not train AI models, and OpenEvidence will delete question and answer data after 24 hours.

Run a global map for about five dollars a month

PMTiles stores a full set of Mapbox Vector Tiles in one file, while HTTP Range requests let browsers fetch only the map areas they need. The article says a laptop load of Europe at city zoom uses about 5MB, costing roughly $0.05 per 1,000 loads on Bunny.net, compared with Mapbox at $5 per 1,000 loads after its free tier. Users can download a planet map from Protomaps or make a smaller one with the command line, then upload it to Bunny.net storage using rclone. The guide also recommends MapLibre GL, the pmtiles npm protocol, and controls for allowed referrers, speed, abuse, and request rates.

Keep human creativity at the center of AI marketing

The article argues that people should lead marketing strategy, ideas, and creative direction, while AI supports the work with research and insights. AI can help prepare briefs from past performance, brand guidelines, market research, and competitor information, but people must decide what the brief means and which ideas work. After people approve a concept, AI can produce versions of its assets and messages for different languages, channels, and audiences. The article says performance data should inform future briefs, and warns that teams should treat its suggested workflow as a flexible guide rather than a fixed rule.

A new method helps multi-agent teams learn from demonstrations

Imitation Bootstrapped Multi-Agent Reinforcement Learning, or IBMARL, combines demonstrations with reinforcement learning for cooperative teams. It uses Round-Robin Behavior Cloning, which lets one person control one agent at a time while the other agents follow their current policies. At each time step, centralized critics choose among actions proposed by imitation and reinforcement learning policies, while a gated penalty guides learning toward imitation actions when the critics value them more. Across four cooperative benchmarks, IBMARL beat imitation policies 2.6 to 4.2 times faster than demonstration-based baselines on two tasks, was the only method to beat them on the other two, and outperformed the demonstrator on two tasks.

GitHub lets teams start AI security scans for 20 repositories

GitHub now lets users start AI Scan for repositories on up to 20 repositories at once. The tool checks default branches for security vulnerabilities, including in languages CodeQL does not support, and can find more issues in languages CodeQL already scans. Results appear in the code scanning backlog alongside CodeQL alerts. The bulk start feature is in public preview and is available on github.com and GitHub Enterprise Cloud with data residency. A GitHub Copilot license is not required.

Nvidia and CoreWeave lead premarket slide in US tech stocks

US tech stocks were set to open lower as Nvidia fell 2.1 per cent in premarket trading and CoreWeave dropped 5.9 per cent. Investors were weighing the rapid growth of AI and its possible effects on the economy. Nvidia’s shares had more than doubled over the past year as demand grew for its graphics processing units and other technologies. The latest decline raised concerns about whether AI growth can continue, along with worries about job losses, economic disruption and greater inequality.

Retailers face new data challenges as AI agents multiply

Retailers and brands deploying AI agents risk creating disconnected systems if they do not improve product data, governance and shared intelligence, according to Akeneo. Its whitepaper describes ten possible agents across the product lifecycle and warns that adding agents can magnify problems in fragmented or unstructured data. Akeneo says businesses need trusted product records, shared rules and workflows, human review, and clear records of changes. People will still provide context, judgment and accountability while AI handles repetitive work. As AI increasingly shapes product discovery and shopping, businesses will need reliable product information that both people and AI systems can understand.

Jamaica approves agreement to explore AI infrastructure investment

Jamaica’s government approved a memorandum of understanding with investors to explore building artificial intelligence infrastructure in the country. Prime Minister Dr. Andrew Holness announced the move at the Jamaica Labour Party’s special delegates conference at the Montego Bay Convention Centre in St. James. He said the goal is to help Jamaica become a developer of AI technology and create jobs in the future digital economy. The government’s AI Task Force is advising on how the country can benefit from the technology. Holness did not name the investors or disclose the proposed investment’s value or timeline.

Thomson Reuters CEO sees AI as an opportunity driven by content

Thomson Reuters Chief Executive Steve Hasker says AI is more of an opportunity than a threat when companies execute it well. He says the company will keep investing heavily in AI-driven innovation. Hasker describes Thomson Reuters as a content-driven technology company with AI at its core.

Sources

NOTE:

This news brief was generated using AI technology (including, but not limited to, Google Gemini API, Llama, Grok, and Mistral) from aggregated news articles, with minimal to no human editing/review. It is provided for informational purposes only and may contain inaccuracies or biases. This is not financial, investment, or professional advice. If you have any questions or concerns, please verify all information with the linked original articles in the Sources section below.

Artificial intelligence Nvidia AI market volatility Stock market decline AI health research Apple OpenEvidence HealthKit AI privacy PMTiles Map data delivery AI marketing AI governance Retail product data Reinforcement learning Multi-agent learning GitHub AI Scan Code security scanning AI infrastructure Jamaica

Comments

Loading...