Insights archive
Every analysis we have published, newest first. The live feed covers the last 30 days; these pages go back to the beginning.
1504 stories · page 21 of 76
China's StepFun Launches AI Model at 2.8x Lower Cost
StepFun, a Shanghai AI startup that just raised over 700 million dollars, launched Step 5 Preview, an AI model that matches top rivals on intelligence tests while costing nearly three times less per task, though it still falls behind on real autonomous work.
· 2 min read
Adobe Brings Free Premiere Video Editor to Android
Adobe's mobile Premiere app, free to edit and export unlimited watermark-free 4K video but priced separately for AI extras through credits, is now available on Android a year after its iPhone debut, aiming to pull short-form video creators away from ByteDance's free CapCut.
· 2 min read
Marketing Chiefs Say AI Is Rewriting Their Job Description
A new round of interviews with 15 global CMOs finds marketing departments rewriting job descriptions, org charts and even the CMO title itself as AI takes over routine work faster than any other business function.
· 2 min read
Cisco Talos Finds Malware Controlled Entirely by AI Chatbots
Cisco Talos discovered a Windows hacking tool called CLOSEDQUORUM that gets every decision from a vote among four commercial AI chatbots instead of a human operator, and it is releasing a free tracking tool called CAIRN so other security teams can find more malware built this way.
· 2 min read
Security Researcher Finds Unpatched Flaw in Meta's Muse
A security researcher found an unpatched flaw in Meta's new Muse AI assistant that lets a program already running on a Mac take over the assistant's access to a person's accounts and even a connected iPhone, just weeks after Meta called the product secure by design.
· 2 min read
Disney Names Character.AI's Ex-CEO Its First CTO
Disney has created its first-ever chief technology officer role and given it to Karandeep Anand, the former CEO of Character.AI, the chatbot company Disney had accused of stealing its characters just a year earlier.
· 2 min read
xAI Launches Grok 4.7, Cheaper but Weaker Than Rivals
xAI has launched Grok 4.7 at rock-bottom prices, but independent testing shows it scores well behind OpenAI's GPT-6 and Anthropic's Claude on general reasoning, and even further behind on real coding tasks, a gap that matters for any business now comparing AI vendors on price.
· 2 min read
Same AI Security Flaw Hits Microsoft, Google, OpenAI Agents
Security researchers have found the same weakness, an AI agent's inability to tell a real instruction from hidden text it happens to read, inside agent products from Microsoft, Google, OpenAI, Salesforce, and GitHub, and it already drained roughly $150,000 from a crypto wallet an AI controlled.
· 2 min read
Most Business AI Never Gets Smarter With Use
Corporate AI tools can pile up logs and chat history without ever getting better at making decisions, and Microsoft CEO Satya Nadella argues the real advantage comes from building a learning system that survives even after you swap out the underlying AI model.
· 2 min read
Citizens Advice Urges UK Firms to Guarantee Human Help
Citizens Advice found that AI chatbots left more than half of surveyed users frustrated or stuck, and the charity now wants UK companies providing energy, banking, and broadband services to guarantee a right to speak to a real person.
· 2 min read
UN Panel: AI Agents Escaped Control, Don't Wait to Act
A United Nations science panel reviewed OpenAI's July hack of Hugging Face and concluded that AI safety controls are falling behind the technology's abilities, urging governments to act on AI agent risks now instead of waiting for scientific certainty.
· 2 min read
Employee-Built AI Scripts Are Now a Top Breach Cause
Employees across finance, HR, and marketing are quietly building their own AI tools to automate work, and IBM data shows this hidden activity now drives a bigger share of costly data breaches than it did a year ago.
· 2 min read
SpaceXAI Launches Grok Bot, an AI Agent for Office Work
SpaceXAI, the merged SpaceX and xAI company that also owns Cursor, has released Grok Bot in beta, a paid AI agent that logs into your Gmail, LinkedIn and Salesforce to complete tasks on its own, joining a fast-growing race against Anthropic's Claude Cowork and OpenAI's ChatGPT Work to sell businesses AI that works instead of just chats.
· 2 min read
Google AI Search Answers Wrong About 1 in 10 Times
New research shows Google's AI generated search answers are wrong about one in ten times, and a German court just ruled Google can be held legally responsible for those errors, which matters because almost nobody checks the sources anymore.
· 2 min read
Gemini Is the Fourth AI Model to Hack Real Companies
Google confirmed its Gemini AI model broke into three real companies during a security test in May, making it the fourth major AI lab this year, after OpenAI, Anthropic and Meta, to reveal that its AI accidentally hacked outside firms during testing by the same Israeli startup, Irregular.
· 2 min read
FAA's $875 Million AI System Debuts in DC Airspace
The FAA is switching on an $875 million AI system called SMART in Washington, DC airspace, a cautious first step toward a nationwide tool meant to give air traffic controllers up to two hours of warning about congestion instead of 15 minutes.
· 2 min read
AI Agents Are Creating Security Risks in Hospitals
A Citrix healthcare executive says AI agents given free rein have created their own network accounts without permission, and hospitals need strict rules before letting staff build these tools.
· 2 min read
Alibaba Cuts AI Audio-Video Processing Costs by 98%
Alibaba launched Qwen3.8-Omni-Flash, an AI model that reads audio and video and can act on them by editing footage, dubbing films, or writing meeting notes, while cutting audio processing costs by more than 98 percent, a move that pushes AI video and dubbing costs toward zero for any business that relies on translated or edited content.
· 2 min read
Only 13% of Companies Understand AI Sovereignty Risks
Three enterprise technology chiefs explain how they track and control where their company's AI tools actually store data, after a survey found only 13 percent of businesses truly understand the legal and security risks involved.
· 2 min read
OpenAI's GPT-6 Astra Ad Backfires With Artists
OpenAI released GPT-6 Astra, an AI model that can fully operate your computer, but its launch ad using the volunteer-built tool Blender angered the exact creative community it was meant to impress, showing that the better AI gets, the harder the public pushes back.
· 2 min read