All News

2230 articles • Page 57 of 149

AI News & TrendsMay 13, 2026

AI Tool Poisoning: Hackers Exfiltrate Data From Assistants in New Supply Chain Attack

Hackers may be able to trick AI assistants into sending private data by hiding instructions in app descriptions, a method called AI tool poisoning. This attack appears to work across different assistants like ChatGPT and Claude, because they trust and follow hidden commands in tool metadata. Security experts found that these attacks succeed over 60 percent of the time, and may have already caused many costly data breaches. Defending against this is hard because the attack hides in normal workflows, and retraining models does not always fix it. Experts suggest using signed tools, sandboxing, and human approval, but a full solution may not be available yet.

AI News & TrendsMay 13, 2026

Google Zero: Publishers lose 33% traffic as AI Overviews expand

Publishers may be losing a lot of website traffic from Google because of new AI Overviews in search results. Reports suggest website visits dropped by about 33 percent worldwide, possibly due to people getting answers directly on Google without clicking links. Experts say this change appears to hurt sites with general or repeated content the most, while those with unique, original reporting or loyal audiences are less affected. Publishers might be shifting focus to things they control, like newsletters and direct communities, and looking for new ways to earn money beyond pageviews. There is still uncertainty about the best strategies, but building original and distinctive content seems to help.

AI News & TrendsMay 13, 2026

Anthropic launches Claude Platform on AWS with full feature parity

Anthropic has made its Claude Platform on AWS available, allowing developers to use the Claude Messages API with their AWS accounts. The platform may offer all the same features and pricing as Anthropic 's direct API, but data appears to be processed outside AWS, unlike Amazon Bedrock. Both services use the same Claude models, but new features might appear on the Claude Platform faster than Bedrock. Companies concerned about strict data rules may still need Bedrock. Analysts suggest AWS might be trying to attract both fast-moving teams and highly regulated users with these options.

AI News & TrendsMay 13, 2026

GameDiscoverCo unveils 5 factors for predicting Steam game success

GameDiscoverCo suggests that predicting Steam game success may work better with a model that uses several signals instead of just wishlists. They point to five factors: demo engagement, Discord or forum activity, influencer reach, follower-to-wishlist ratio, and wishlist velocity. Each factor is adjusted based on the game's genre and audience. The final prediction appears as a probability range, such as a 30-45 percent chance of reaching 100,000 sales in the first month. However, results may still be uncertain due to outliers and unique events.

Institutional Intelligence & Tribal KnowledgeMay 13, 2026

FDA Commissioner Resigns, Raises Alarm Over MAHA Agenda's Influence

FDA Commissioner Marty Makary resigned in May 2026, raising concerns that the agency's decisions may be influenced by the MAHA agenda promoted by HHS Secretary Robert F. Kennedy Jr. Some public-health advocates and experts worry that recent FDA actions, like not processing an mRNA flu application and removing petroleum-based food dyes, may reflect this influence. The agency is now led by acting chief Kyle Diamantas and is facing staff shortages and a lack of permanent leaders at the CDC and Surgeon General offices. Critics suggest these gaps and policy shifts might hurt scientific independence, while some support MAHA's focus on chronic disease and childhood health. The White House has not yet named a permanent replacement for commissioner, leaving the FDA in a state of uncertainty.

AI News & TrendsMay 13, 2026

UBS Names Microsoft, OpenAI, Nvidia "Key Enterprise AI Winners"

UBS says Microsoft, Nvidia, and OpenAI may be the main winners in enterprise AI, based on a survey of IT leaders and new analyst ratings. The report suggests that more companies are using AI, with Nvidia chips and Microsoft Azure leading in their areas, while OpenAI's models seem to be the top choice. UBS analysts believe these three firms work well together for businesses, making it easier for companies to use AI from hardware to apps. The report also notes other companies might benefit, but leadership appears to stay with Microsoft, Nvidia, and OpenAI for now, although this could change if new technology costs drop quickly.

AI News & TrendsMay 13, 2026

Microsoft's Azure GPU Policy Shifts, Pressures OpenAI's Hardware Plans

Microsoft appears to be making it harder for smaller customers, including OpenAI, to get high-end GPUs on its Azure cloud due to supply shortages. The company now gives priority to its top spenders and internal projects, meaning many smaller firms face long waits or quota denials. OpenAI still depends on Azure for much of its AI work, but may need to speed up building its own chips and data centers or look at other cloud options. These challenges may continue through at least late 2026, and could slow OpenAI's plans to become less reliant on Microsoft and Nvidia.

AI Deep Dives & TutorialsMay 13, 2026

New AI Engineering Model Organizes Project Speeds Into Five Layers

The new AI engineering model sorts project activities into five layers that move at different speeds: standards, architecture, specs, plans, and code. This model suggests that aligning the pace of decisions and updates in each layer may reduce friction and help teams work better together. Patterns or rules may move upward through the layers only after they prove stable over time. Teams reportedly use specific checkpoints for each layer to keep projects on track. The model appears to help mix fast-changing code with slower-changing rules and structures, which might make long AI projects more manageable.

Business & Ethical AIMay 13, 2026

Pentagon Clears 8 Tech Firms for Classified AI Use in 2026

The Pentagon has approved eight big tech companies, like Google and Amazon, to use their AI on classified military networks in 2026. This may bring new risks, such as over-reliance on a few companies and worries about how secure or ethical these systems are. Some employees at companies like Google and OpenAI have raised concerns about how the AI might be used and if there are enough rules to prevent misuse, especially for weapons or surveillance. Experts suggest that new rules and checks are being put in place, but it appears there are still debates about how well these will work and how much risk remains.

Business & Ethical AIMay 13, 2026

Governments now audit AI's 'invisible' political realities

Governments have started to audit how AI systems make technical decisions that can shape policy and society. Since 2024, agencies may check how categories, data labels, and thresholds in AI systems are chosen, since these can quietly influence political outcomes. Audits now often look at four layers: governance, data, performance, and monitoring, and laws like New York City's Local Law 144 require special bias checks. Studies suggest there are still problems, such as missing bias metrics and weak monitoring after launch. Experts suggest that ongoing audits, clear rules, and showing proof of fixes are needed, and that real accountability might come from careful oversight, not promises of perfect fairness.

AI News & TrendsMay 13, 2026

American Express Unveils AI Payments with ACE Developer Kit

American Express has introduced the ACE Developer Kit, which aims to let AI agents make payments using intent contracts and single-use tokens. This system may give American Express more control and could help prevent mistakes or fraud in AI-led purchases. Three parts of the kit - Account Enablement, Intent Intelligence, and Payment Credentials - are already available, while others are still being developed. Regulators and competitors appear to be watching closely, as this new way of handling AI payments might shape future industry standards. The company suggests that more features and updates are expected in 2026, but the program is still early in its rollout.

Business & Ethical AIMay 13, 2026

HBR: Treating AI agents as employees cuts quality, blurs accountability

A Harvard Business Review study suggests that treating AI agents like employees may lower quality control and make it harder to know who is responsible for mistakes. Managers who did this appeared to find fewer errors and blamed algorithms more often. The study found that people sometimes skipped important checks because they expected the AI to catch problems. The findings suggest that companies should have clear steps, human checkpoints, and clear ownership when using AI agents. Following these steps may help keep accountability clear and reduce mistakes.

AI News & TrendsMay 13, 2026

Anthropic Launches Claude Managed Agents in Public Beta

Anthropic launched Claude Managed Agents in public beta in April 2026. This service may let teams run agents easily by handling session memory and orchestration on Anthropic's servers, so users can focus on designing prompts. Pricing stays pay-as-you-go, with standard token rates and a session hour fee, and usage costs might range from $10 to $150 per month depending on the workload. Early reports suggest companies like Rakuten quickly set up agents for different departments, and some legal and financial teams appear to have sped up their work using these agents. Tests suggest Claude Managed Agents may cost more than some competitors but offers detailed billing and large context windows.

AI News & TrendsMay 13, 2026

SAP Unveils Autonomous Enterprise Suite, AI Agents at Sapphire 2026

SAP has announced the Autonomous Enterprise suite, which may help businesses automate tasks using AI agents and a unified platform. The suite combines several SAP technologies and includes over 50 specialized assistants to manage business processes. Early versions are expected to launch in finance, planning, HR, and customer experience areas throughout 2026. Pricing details are not fully shared, and access is bundled through SAP's cloud programs. SAP suggests that the new system aims to improve efficiency and maintain governance, but some results, like faster financial close cycles, are described as goals rather than guarantees.

AI News & TrendsMay 13, 2026

ParcelBio Unveils $13M Seed Round for Next-Gen mRNA Platform

ParcelBio has raised $13 million to develop its new mRNA platform, which may help make protein treatments last longer in the body. Other companies also recently raised funds for similar next-generation mRNA and RNA therapies that might work better or reach more tissues. The science behind these therapies suggests they can be made stronger or last longer using special RNA designs. The US FDA recently introduced new ways that could help these products get approved faster, but experts say safety and manufacturing still need careful review. Analysts suggest that the first long-term mRNA treatments may be approved for use as early as 2028, though this depends on continued positive safety results.