DeepMind's Co-Founder Has a Warning for Remote Workers
Google’s AI chief scientist just said the quiet part out loud: if your job can be done from home on a computer, it might not be your job for long. The holiday week is wrapping up with a reality check about what AI disruption actually looks like.
📰 The Rundown
🏠 DeepMind Co-Founder: Remote Jobs Are First on the Chopping Block
➡️ The move: Google DeepMind co-founder Shane Legg warned in a new interview that many remote, computer-based jobs could disappear over the next decade. His rule of thumb is simple: “If you can do the job remotely over the internet just using a computer, then that job is potentially at risk.” Digit.in
⚡ Why it matters: Legg isn’t some outsider making predictions. He’s chief scientist at the lab that created AlphaFold and Gemini. His view is that AI’s current weaknesses in reasoning and learning “will all get addressed” over the coming years. The implication: software engineering teams that needed 100 people might soon need 20 who use advanced AI tools effectively.
🎯 Your takeaway: The work-from-home revolution created millions of jobs defined by “screen time equals productivity.” That equation is changing. The professionals who thrive won’t be the ones who resist AI, they’ll be the ones who use it to multiply their output.
🎁 Claude Doubles Usage Limits Through New Year’s Eve
➡️ The move: Anthropic is giving Pro and Max subscribers a holiday gift: 2x usage limits from December 25 through December 31. Both five-hour limits and weekly caps are doubled. This applies to Claude.ai, Claude Code, and Claude in Chrome.
⚡ Why it matters: If you’ve been bumping against Claude’s rate limits while working on year-end projects, you just got runway. This is Anthropic’s way of letting power users stress-test how much AI assistance they actually want in their workflow.
🎯 Your takeaway: Use this week to experiment with heavier Claude usage than normal. Notice when you hit friction and when you find flow. The data will inform whether a higher tier makes sense for your 2026 workflow.
📊 Gemini Triples Market Share as ChatGPT’s Lead Shrinks
➡️ The move: New data from Similarweb shows Google’s Gemini has surged from 5.4% to 18.2% of generative AI web traffic over the past year. That growth came directly from ChatGPT, which dropped from 87.2% to 68.0%. Meanwhile, Microsoft’s Copilot remains stuck at just 1.2% despite being embedded in Windows and Edge. Benzinga
⚡ Why it matters: Google’s advantage is native distribution. Gemini is built into Chrome, Android, Workspace, and Search. Users don’t need to navigate to a separate app; the AI meets them where they already work. Copilot’s stagnation proves that pre-installation alone isn’t enough. Product trust matters.
🎯 Your takeaway: The AI tool you use matters less than how deeply you integrate it into your workflow. Whether you’re team ChatGPT, Claude, or Gemini, the professionals pulling ahead are the ones who’ve made AI ambient rather than occasional.
🔧 Tool Spotlight: Flick
Flick is a social media management platform with Iris, a built-in AI content assistant that helps you brainstorm, draft, and polish posts across all your channels. Instead of staring at a blank caption box, Iris generates post ideas from a topic, expands them into full drafts, and repurposes content for different platforms automatically.
What makes it different: Most AI writing tools give you generic output. Iris learns your brand voice and generates content that actually sounds like you. It also includes scheduling tools so you can plan your entire social calendar without switching apps.
Best for: Marketing managers, small business owners, and anyone responsible for keeping social channels active who wants to cut content creation time without sacrificing quality.
Pricing: Free 7-day trial. Paid plans start at £14/month for individuals, with team plans available.
👉 Try it: flick.social
✨ Try This Today: The Briefing Document Technique
Every time you start a new AI conversation about an ongoing project, you probably spend the first few minutes explaining context: what the project is, who the stakeholders are, what’s already been decided. It’s tedious, and you often forget key details that change the quality of the output.
The technique: Create a short briefing document (200-400 words) that you paste at the start of any related AI conversation. You’ve just given AI the equivalent of a week of onboarding in 30 seconds.
What to include:
1. Project overview (2-3 sentences on what, why, and core goal)
2. Key stakeholders (who cares and what they prioritize)
3. Constraints (budget, timeline, technical requirements)
4. Decisions already made (what’s off the table)
5. Success criteria (what “done” looks like)
Why it works: You front-load context once and reuse it indefinitely. Every AI session for that project starts at the same baseline. No more re-explaining, no more inconsistent outputs when you forget to mention something.
Time required: 15-20 minutes to write once. Saves 5-10 minutes per AI session after that.
📚 Go deeper: The Briefing Document Technique: Full Lesson
✨ The Wire
🔗 Google fired procurement executives after AI memory chip supply shortages disrupted its infrastructure plans, underscoring how critical the chip supply chain has become for AI development. Digit.in
🔗 OpenAI’s GPT-5.1 is now powering ChatGPT according to Zapier’s 2025 productivity tool roundup, though some users report the latest release feels less pleasant to interact with than previous versions. Zapier
🔗 Warp Terminal launched Agents 3.0, transforming from a smart terminal into a full agentic development studio that can run live debugging and database queries under your supervision. Medium
Neural Notes — AI that amplifies your value, not replaces it.
