19-year-old founder emerges from stealth with an AI-first personal computer
Zain Javaid’s Ghost AI launched Core, a personal computer that offers 24/7 AI features, powered by on-device models. Core can complete tasks across connected apps and devices, with the computer remembering key details about your life over time. It joins a growing wave of AI-first hardware that includes the Googlebook, new Mac Mini, and Surface Laptop Ultra. Watch how Core works.
Nolla Health begins testing one of the first end-to-end AI doctors
Several AI tools can renew prescriptions that a human doctor wrote. Nolla Health has received regulatory approval to go one step further: handling the entire medical process. A dermatology-focused pilot program is rolling out in Utah that will let the AI assess symptoms, prescribe treatment, and follow up with patients as human doctors stay on call in the background.
New York-based lab expands the Western open frontier
Reflection previewed Beam, its first open weight model built for coding, reasoning, and agentic workloads. The lab claims Beam’s performance matches China’s GLM 5.2 on advanced reasoning, while using 3-4x less compute. This release puts Beam in a small-but-growing group of Western labs with open models, along with Meta, Nvidia, and Thinking Machines.
How to create a recurring security-news brief for your IT team with Grok Bot
Step 1: Open Grok Bot and create a new Bot for cybersecurity monitoring.
Step 2: Tell it which topics to track, such as ransomware, zero-days, major vulnerabilities, phishing, cloud security, and data breaches.
Step 3: Tell Grok Bot to prioritize CISA, vendor advisories, security researchers, and reputable cybersecurity publications over unverified posts.
Sample Prompt: “Review the latest cybersecurity developments relevant to [company/industry]. Prioritize vulnerabilities, breaches, ransomware, phishing campaigns, and vendor advisories affecting [tools/vendors]. For each item, explain what happened, why it matters, severity, affected products, recommended action, and add a source link.”
- Run the brief once manually and refine the topics, sources, and vendors based on what creates useful signal versus noise.
- Save the workflow as a Skill, then create a Routine to run it every weekday morning or on your preferred schedule.
- Format each brief into ‘Critical Now’, ‘Watchlist’, ‘Recommended Actions’, and ‘Source Links; so the IT team can scan it quickly.
Who’s shaping the AI conversation this week? Probably not the people you expect
CEOs and researchers usually dominate AI’s headlines. But so far this week, two unlikely people have stolen the spotlight: the Pope and a YouTuber.
The Pope says AI lacks the “spark of humanity”
In a post that generated nearly 25M views, Pope Leo XIV said it’s “becoming urgent” to distinguish human art from machine-generated art. It’s a bit unexpected for the leader of the Catholic Church to continue weighing in on tech policy, but the situation makes more sense in light of a recent NYT report.
Per the NYT, Anthropic co-founder Chris Olah lobbied the Pope’s advisors to consider the possibility that Claude may be conscious, which may have prompted the Pope’s statement. The AI lab hasn’t publicly argued that Claude is conscious, but it has hinted at the idea at least once.
Meanwhile, YouTuber PewDiePie came out of retirement to release a custom AI model
PewDiePie posted a farewell video called “Thank you and goodbye.” Two weeks later, he returned to announce Ajax, his own uncensored open weight model. The full video offers a great introduction to open weight AI, while also telling the full story of why OpenAI banned him twice.
Plan a team communication reset
Prompt: Act as an experienced organizational communications strategist and team effectiveness advisor. Create a practical communication reset plan for ……………