AI News

Follow the latest AI and automation news

22 May 20266 Dhuʻl-Hijjah 1447 AH

FTC to Require Cox Media Group, Two Other Firms to Pay Nearly $1 Million to Settle Charges They Deceived Customers About “Active Listening” AI-Powered Marketing Service

The FTC announced that Cox Media Group and two other firms will pay nearly $1 million to settle charges related to deceptive marketing practices. In 2024, Cox attempted to sell advertising packages based on 'active listening', claiming smart devices captured real-time consumer intent data. However, complaints revealed that the service did not actually listen to conversations or utilize voice data. Instead, it involved reselling email lists acquired from other data brokers at a significant markup.

Models

22 May 20266 Dhuʻl-Hijjah 1447 AH

Chinese AI startup Zhipu rolls out ultra-fast version of flagship model

Zhipu, a Chinese AI startup, has launched its latest flagship model, GLM-5.1, achieving an impressive output speed of 400 tokens per second. This model uniquely combines flagship capabilities with ultra-low latency, marking a significant advancement among domestic large models. This development positions Zhipu at the forefront of the AI market, opening new opportunities for developers and users in the local AI landscape.

Governance

22 May 20266 Dhuʻl-Hijjah 1447 AH

Can OpenAI’s ‘Master of Disaster’ Fix AI’s Reputation Crisis?

Chris Lehane, OpenAI's global affairs chief, is urging a calmer debate on the societal impacts of AI. He aims to encourage states to enact laws that support OpenAI's growth rather than hinder it. This call comes amid rising concerns about AI's implications for jobs and privacy. Lehane believes the current discourse should focus more on the potential benefits of AI technology rather than solely on its risks. He emphasizes that laws must be carefully crafted to avoid negative impacts on innovation, highlighting the need for collaboration between governments and companies.

💡

How Virgin Atlantic ships faster with Codex

How can AI enhance app development in other companies?

Agents

OpenAI named a Leader in enterprise coding agents by Gartner

22 May 20266 Dhuʻl-Hijjah 1447 AH

Agents

22 May 20266 Dhuʻl-Hijjah 1447 AH

OpenAI named a Leader in enterprise coding agents by Gartner

OpenAI has been recognized as a Leader in Gartner's 2026 report on AI coding agents, with Codex identified as an innovative and widely applicable tool for enterprises. These rankings reflect OpenAI's ongoing efforts to advance AI technologies, with Codex representing a significant step towards enhancing programming efficiency in workplace environments. The report indicates that OpenAI has made notable progress in delivering advanced software solutions, emphasizing the role of AI in improving productivity and fostering innovation in programming over the coming years.

Models

Roundtables: Can AI Learn to Understand the World?

21 May 20265 Dhuʻl-Hijjah 1447 AH

Models

21 May 20265 Dhuʻl-Hijjah 1447 AH

Roundtables: Can AI Learn to Understand the World?

The recent session focused on advancements in world models within artificial intelligence, with speakers aiming to overcome the limitations of large language models. Participants, including Matt Honan, discussed how AI can better interact with the physical world. This session was recorded on May 21, 2026, emphasizing the relevance of this topic in current AI discussions. World models are essential for understanding how AI interacts with its environment, paving the way for new applications in everyday life.

Agents

Giving Agents Computers — Ivan Burazin, Daytona

21 May 20265 Dhuʻl-Hijjah 1447 AH

Agents

21 May 20265 Dhuʻl-Hijjah 1447 AH

Giving Agents Computers — Ivan Burazin, Daytona

Daytona has shifted from traditional development environments to AI-specific ones, catering to developers' needs for dynamic and adaptable resources. This change allows clients to access composable computers via APIs, enhancing efficiency in task execution. The company focuses on providing stateful sandboxes that can handle workloads from zero to 100,000 CPUs, reflecting a significant market evolution towards flexible development environments. This innovation is crucial as developers increasingly require robust solutions for complex software engineering tasks.

Agents

Datasette Agent

21 May 20265 Dhuʻl-Hijjah 1447 AH

Agents

21 May 20265 Dhuʻl-Hijjah 1447 AH

Datasette Agent

Datasette has announced the release of Datasette Agent, an AI assistant that provides a conversational interface for querying stored data. This launch marks the culmination of over three years of development on the LLM Python library, integrating LLM with Datasette. Users can enhance their experience by adding the datasette-agent-charts plugin, which generates visual data representations. The announcement includes a demo video showcasing the assistant's capabilities in action.

Business

Starbucks Halts AI Inventory Management Program After Nine Months

21 May 20265 Dhuʻl-Hijjah 1447 AH

Business

21 May 20265 Dhuʻl-Hijjah 1447 AH

Starbucks Halts AI Inventory Management Program After Nine Months

Starbucks has halted its AI-powered inventory management program after just nine months of implementation across its North American stores. The decision was made due to operational issues, including errors in product recognition and inventory counting. The program was intended to enhance product availability and reduce shortages, part of CEO Brian Niccol's strategy. However, employees reported that the system struggled to differentiate between similar products and sometimes failed to detect items during scans.

Agents

Google checks websites for llms.txt in new agentic browsing audit

21 May 20265 Dhuʻl-Hijjah 1447 AH

Agents

21 May 20265 Dhuʻl-Hijjah 1447 AH

Google checks websites for llms.txt in new agentic browsing audit

Google has announced a new test aimed at evaluating how well websites handle AI agents. This is being done through a new experimental category called "Agentic Browsing" in its Lighthouse analysis tool. The test includes checking for the presence of an llms.txt file on targeted websites. This initiative is part of Google's efforts to enhance website performance amid the growing use of AI agents. By conducting this test, Google can assess how prepared websites are to interact with AI technologies, which may influence their ranking in search results.