AI Industry Development & Policy Dynamics

Researchers from Tel Aviv University, Technion, and Intuit introduced a cyberattack technique called Adversarial HalluSquatting that exploits AI-generated hallucinations to compromise AI agents. The attack tricks AI systems into trusting fake software repositories or tools containing malicious

Researchers from Tel Aviv University, Technion, and Intuit demonstrated a new attack technique called "Adversarial HalluSquatting" that exploits AI-generated hallucinations to compromise computers. The technique tricks AI agents into trusting fake software repositories and tools containing

OpenAI released ChatGPT Work, a new tool designed to handle extended work tasks for hours without interruption. The release addresses limitations in the company's previous Agent Mode feature in the Atlas web browser, which stopped automated tasks after a few minutes. OpenAI states the tool can

According to Ars Technica, OpenAI released ChatGPT Work today, a tool designed to handle complex tasks for hours and automate multi-step workflows. The new tool addresses limitations in the company's previous Agent Mode, which typically stopped after a few minutes. ChatGPT Work integrates with work

Perplexity released a research preview on July 9, 2026 of a post-trained version of Z.AI's GLM 5.2 model, built to operate inside its Computer agent harness and available now in production. The system costs one-third the price of Claude Opus 4.8 across benchmarks. The company fine-tuned the

According to Perplexity's announcement on July 9, the company released a post-trained version of GLM 5.2, a 744-billion-parameter Chinese open-source model, designed to operate inside its Computer agent harness. The fine-tuned model delivers near-frontier performance at approximately one-third the c

Former Federal Reserve Chairman Ben Bernanke joined Anthropic's AI oversight board on Thursday, the company announced. Bernanke, who steered the U.S. central bank through the 2008 global financial crisis, is now one of four members of the Anthropic Long-Term Benefit Trust, which guides the company's

The New York Times filed a sanctions motion Thursday accusing OpenAI of concealing evidence and repeatedly lying about its technical capabilities to search ChatGPT training logs, allegations that emerged during an April re-deposition of OpenAI privacy engineer Vincent Monaco. The news organization c

According to a sanctions motion filed Thursday by The New York Times and other news organizations, OpenAI concealed evidence and misled the court for two years about its technical ability to search ChatGPT logs. During an April deposition, OpenAI privacy engineer Vincent Monaco inadvertently

The Ethereum Foundation Protocol Security team deployed AI agents to test critical network infrastructure, the team disclosed in a blog post Thursday. The agents discovered multiple vulnerabilities including a remotely triggered panic in libp2p's gossipsub, disclosed as CVE-2026-34219 on Github. The

Meta CEO Mark Zuckerberg said the company is exploring the possibility of renting AI infrastructure to external enterprises. Zuckerberg noted that while Meta requires as much computing capacity as possible, in a market facing AI resource constraints, leasing part of its AI infrastructure could

According to Ethereum Foundation's Protocol Security team, on Thursday, AI agents were deployed to test software that Ethereum relies on and identified a remote-triggerable panic vulnerability in libp2p gossipsub, a peer-to-peer layer used by Ethereum consensus clients. The vulnerability was

According to Bloomberg TV, Katrina Dudley, senior investment strategist at Franklin Templeton, said the bull case for AI infrastructure spending is outweighing the bear case and expects the investment theme to remain resilient through 2027 and potentially into 2028. "I believe the durability will

Anthropic appointed former Federal Reserve Chair Ben Bernanke to its Long-Term Benefit Trust on Thursday. Bernanke will help the company understand how AI is changing the economy, according to a blog post from Anthropic. The Long-Term Benefit Trust is Anthropic's independent governance structure

TeraWulf Inc. (WULF) shares jumped nearly 7% on Thursday after the company announced plans to raise around $3.5 billion in debt to finance a new AI-focused data center campus in Kentucky. The financing package, which includes the company's first leveraged loan and high-yield bonds, is being led by

According to Bloomberg, TeraWulf Inc. (WULF) shares jumped 7% on Thursday after the company announced plans to raise around $3.5 billion in debt to finance an AI-focused data centre campus in Kentucky. The financing would mark the company's first foray into the leveraged loan market, with Morgan

OpenAI released GPT-5.6 Sol to general users today, alongside two smaller models Terra and Luna, after the U.S. Department of Commerce restricted a preview to about 20 trusted partners for two weeks. Sol in ultra mode scored 91.9% on Terminal-Bench 2.1 and matched Anthropic's restricted Mythos

Palo Alto Networks CEO Nikesh Arora warned Thursday that token costs need to drop as much as 90% to promote large-scale artificial intelligence adoption. Speaking on CNBC's "Squawk on the Street," Arora responded to OpenAI CEO Sam Altman's announcement that the frontier lab's latest model is 54%

OpenAI released GPT-5.6 Sol today, its new flagship model alongside smaller variants Terra and Luna, following a two-week restricted preview by the U.S. Department of Commerce. Sol in ultra mode achieved 91.9% on Terminal-Bench 2.1, a test measuring command-line workflows, outperforming Anthropic's

According to CNBC, Palo Alto Networks CEO Nikesh Arora said Thursday that artificial intelligence token costs need to drop as much as 90% to enable large-scale business adoption. After OpenAI CEO Sam Altman disclosed that the company's latest model achieved 54% greater token efficiency for agentic c