SpaceXAI on September 21, 2026 released Grok 4.7, its newest model for coding and knowledge work, priced from $2 per million input tokens and $6 per million output tokens. In the release announcement, ...

A widespread outage has taken out popular AI platforms including ChatGPT, Claude, and Grok. Update 12:38pm PT: All services are back to normal across ChatGPT, Claude, and Grok. Update 8:49am PT: ...

Update: Service appears to be coming back online for some after about half an hour. Your mileage may vary. It’s not just you. Claude, ChatGPT/Codex, and Grok are currently experiencing what appears to ...

X users are reporting that Grok is currently down. There are over 1,000 reports, and the count is growing. Update: September 3, 2026 (2:55 PM ET): The number of reports has dropped significantly, ...

Gemini for Government, or GenAI.mil, is the War Department's new artificial intelligence tool. (Graphic: DOW screenshot) The Department of War has launched Starshield AI’s Grok for Government and ...

The United States Department of Defense just added special versions of ChatGPT and Grok to its own internal software network. This big move gives millions of military and civilian workers new ...

The Pentagon announced on Monday that it brought on Starshield AI’s Grok for Government and OpenAI’s ChatGPT for unclassified military use, expanding the number of AI assistants on the department’s ...

A beginner-friendly walkthrough of Grok Imagine 1.0, covering the key features, how to get started, and how to create content ...

An AI-generated image (Grok) from the prompt (August 29, 2026): “Make an image of how you see yourself, Grok.” Grok, the AI assistant from xAI, is still one of the most useful AI tools on Apple ...

Adversa AI researcher Rony Utevsky devised a new attack technique, called Cryptographic Context Injection, that bypasses AI safety filters by sending instructions as AES-encrypted ciphertext and ...

People who used Grok Lite this week were no doubt surprised when the bot started sending long, large paragraphs of total gibberish. This is only slightly worse than its usual output, but it still had ...