OpenAI to Cut Revenue Share

Welcome to another edition of AI PlanetX.

OpenAI lowers Microsoft’s revenue share; xAI sheds more than 500 employees; Apple AI veteran behind Siri steps down.

• Hottest AI News

• Top AI & SaaS Tools

• AI Tutorial: Generate Stunning Research Reports

• Top AI & Tech News

• AI Art Spotlight

• Resource of the Day: Writing Effective Tools for Agents

• AI Video Tutorial

• Course of the Day: Making Friends with AI

Hottest AI News

OpenAI

OpenAI Plans to Cut Microsoft Revenue Share from 20% to 8%

OpenAI is set to dramatically reduce its revenue-sharing obligations with Microsoft over the next six years, potentially keeping tens of billions more dollars for itself as the AI giant restructures its business model.

Details:

• OpenAI plans to cut partner revenue share from 20% to 8% by decade’s end, according to The Information. Microsoft, its biggest partner, would take the largest hit, while OpenAI could keep over $50 billion that would have gone to partners. The report didn’t say if that’s annual or cumulative

• The two companies are also renegotiating server rental costs, adding tension to an already shifting partnership. Talks reflect OpenAI’s rapid growth and rising market value

• This shake-up aligns with OpenAI’s broader corporate shift. A non-binding deal with Microsoft sets terms for its for-profit transition. The nonprofit arm would get over $100 billion—about 20% of a $500 billion valuation—making it one of the richest nonprofits in the world

The revenue shift shows OpenAI’s confidence and push for financial independence, but its success hinges on hitting lofty valuation goals.

Master ChatGPT for Work Success

ChatGPT is revolutionizing how we work, but most people barely scratch the surface. Subscribe to Mindstream for free and unlock 5 essential resources including templates, workflows, and expert strategies for 2025. Whether you're writing emails, analyzing data, or streamlining tasks, this bundle shows you exactly how to save hours every week.

Subscribe for Your Free Bundle

xAI

xAI Fires 500+ Workers Training Grok in Friday Night Layoff Massacre

Elon Musk's xAI has executed a massive layoff of its Grok training team, cutting at least 500 workers in a sudden Friday night purge that decimated the company's largest department.

Details:

• xAI laid off its data annotation team—the workers who trained Grok to understand the world. In an email, the company said it was shifting from generalist tutors to specialist roles, cutting most of the generalist positions that built Grok’s foundation

• Workers got less than 24 hours’ notice after a chaotic Thursday night test. Senior staff had their Slack logins cut earlier in the week. Employees were pulled into one-on-one reviews and asked to flag top performers. That night, team lead Diego Pasini announced surprise reorganization exams on coding, STEM, memes, and Grok’s personality

• The layoffs were executed brutally. Logins were revoked despite promises of continued pay. Friday night emails confirmed jobs were ending, with salaries covered through contracts or November 30. Slack membership fell from 1,500 to just over 1,000 in hours. At the same time, xAI posted job listings, claiming it would expand specialist tutor roles by “10X”

The irony is sharp: xAI cut the very people who taught Grok to act human while framing it as growth. The layoffs raise doubts about Grok’s future and whether the promised 10X hiring is real or just cost-cutting in disguise.

Top AI & SaaS Tools

• Claritee (Life-time Deal): Build websites, apps, and landing pages with AI that turns simple prompts into sitemaps, layouts, UI, and real content

• Jobright: Locate jobs that fit your skills, tailor your resume for each position, and submit applications on your behalf automatically [F-R-E-E]

• Google AI Edge: A gallery that showcases on-device ML/GenAI use cases and lets people try and use models locally [F-R-E-E]

• Jules: Create custom courses on any topic in seconds from a simple prompt—available in nine formats, including text, audio, games, and quizzes

• SciSpace Agent: AI agent that automates research with 150+ academic tools, acting as your Co-Scientist to read literature, analyze data, and draft papers [F-R-E-E]

Daily News for Curious Minds

Be the smartest person in the room by reading 1440! Dive into 1440, where 4 million Americans find their daily, fact-based news fix. We navigate through 100+ sources to deliver a comprehensive roundup from every corner of the internet – politics, global events, business, and culture, all in a quick, 5-minute newsletter. It's completely free and devoid of bias or political influence, ensuring you get the facts straight. Subscribe to 1440 today.

Sign up now!

AI Tutorial

How to Generate Stunning Research Reports With Kimi

Kimi just raised the bar for research. From a single prompt, Kimi Researcher generates professional-quality reports in minutes, blending deep analysis with interactive graphics—without the limits found in other platforms.

1. Go to Kimi

   Open your browser, go to Kimi, and sign up or log in with your existing credentials.

2. Select Researcher tool

   From the dashboard, choose the Researcher button to open the input field.

3. Enter research request

   Be clear and specific. For example:

   “Investigate the rise of remote work since 2020. Track major corporate policy shifts, adoption by different industries, productivity studies, employee satisfaction surveys, and the impact on commercial real estate markets in New York, London, and San Francisco.”

4. Answer clarifying questions

   Kimi may ask for details such as:

   • Global vs regional scope

   • Current trends vs long-term forecasts

   • Social, economic, or policy focus

5. Wait while Kimi researches

   The AI runs multi-turn searches and reasoning, which may take a few minutes.

6. Review generated report

   You’ll get a detailed text summary plus an interactive report.

7. Explore visuals

   Check the included charts, graphs, timelines, and dashboards to see data from different angles.

8. Download results

   • Use the download icon to save a PDF version

   • Keep it for sharing, offline use, or citation

9. Copy interactive code (optional)

   If you need to embed the report into a website, presentation, or workspace, copy the provided code.

Note: Start with smaller, focused research requests to see how the tool responds. Then scale up to multi-part, complex prompts once you’re comfortable with the workflow.

Top AI & Tech News

• Robby Walker, a senior Apple AI executive who led the Answers, Information and Knowledge team and previously oversaw Siri, is leaving Apple

• Penske Media, owner of Rolling Stone, Billboard, and Variety, sued Google for allegedly using publishers’ content to power its AI Overviews

• Tucker Carlson claimed ex-OpenAI researcher Suchir Balaji was “definitely murdered” during an interview with CEO Sam Altman

AI Art Spotlight

Model: Midjourney V7

Prompt:

A realistic blurred photo of a rainy street with glowing streetlights reflected in puddles. In the foreground, a glowing neon line art of a girl holding an umbrella, smiling as she looks up at the sky. She’s outlined in thick glowing white lines. Above the umbrella, sketchy doodles of stars and raindrops float softly. Warm doodles add a healing touch. Her glowing shoes lightly touch the wet pavement. --ar 9:16 --raw --v 7

Resource of the Day

Writing Effective Tools for Agents—with Agents, by Anthropic

This guide explains how to design, evaluate, and refine tools that LLM agents use, emphasizing that effective agent behavior depends on well-crafted tools and specs.

It walks through a practical workflow: prototype tools (ideally via MCP servers or desktop extensions), create realistic, multi-step evaluations, run programmatic agent loops with metrics, and iterate using agents like Claude Code to analyze transcripts and improve implementations.

Top AI Video Tutorial

Nano Banana vs. Photoshop: A Fair Comparison

Complimentary AI Course of the Day

Making Friends with AI [Taught by Google’s Former Chief Decision Scientist]

This course explains how machine learning works and how to use it responsibly. Aimed at both beginners and those with some background, it focuses on concepts over math. You’ll learn what ML is, why it’s useful, how it’s applied, how to frame real-world problems, the role of human judgment, and key ethical issues.