AI Cheating Case Hits Brown Univ

OpenAI Safety Head Exits

In partnership with

AI PlanetX

Welcome to another edition of AI PlanetX.

Brown suspects widespread AI cheating; OpenAI revamps safety; Samsung pushes AI data consent.

Inside This Edition: đź’Ž

  • Hottest AI News

  • Top AI & SaaS Tools

  • AI Tutorial: Build Full-Stack CRM Using Antigravity

  • Top AI & Tech News

  • AI Art Spotlight

  • Prompt of the Day: 99 Claude Prompt Commands

  • Featured AI Video

Hottest AI News

AI in Education

Brown Professor Suspects Most Students Used AI to Cheat on the Midterm

Brown economics professor Roberto Serrano suspects most of his 86 students used AI on a take-home midterm allowed after campus shooting anxiety. The unusually high results triggered an academic integrity crisis at Brown.

Details:

  • The midterm average hit 96%, far above the course’s historic 65–80% range, though the test was harder. ChatGPT produced similarly convoluted answers, including the same unusual proof

  • Serrano moved the final in person. Eighteen students dropped, nine skipped it, the average fell to 48.6%, three earned zero, and 19 failed. He voided the midterm and made the final worth 80%

  • Serrano called Brown’s response “meek” after being told to file individual complaints. Brown said procedures apply. Its report urged clearer codes, dialogue, enforcement and less punishment

The case shows universities need stronger AI rules, redesigned assessments, resources, and honest use that protects real learning—not just punishment.

100+ Claude Code hacks to ship code 10X faster

Top engineers at Anthropic and OpenAI say AI now writes 100% of their code.

If you're not using AI, you're spending 40 hours doing what they do in 4.

These 100+ Claude Code hacks fix that and help you ship 10x faster.

Sign up for The Code and get:

  • 100+ Claude Code hacks used by top engineers — free

  • The Code newsletter — learn the latest AI tools, tips, and skills to code faster with AI in 5 minutes a day

OpenAI

OpenAI Safety Leader Departs as Research and Safety Teams Are Reshaped

OpenAI’s head of safety systems, Johannes Heidecke, is leaving as the company reorganizes safety around frontier-model research, faster training cycles, shorter release windows, and increasingly capable AI.

Details:

  • Safety work reports to Mia Glaese in her broader role as VP of research and safety. Saachi Jain becomes interim safety systems chief as OpenAI coordinates research and releases more closely

  • Heidecke joined OpenAI in 2021 and became safety leader in 2024 after Lilian Weng left. Mark Chen noted faster model training and shorter release cycles made safety coordination harder

  • His departure follows GPT-5.6’s launch and OpenAI’s warnings about concerning misaligned behavior. Safety researcher Joshua Achiam is also leaving after nine years

Fidji Simo is stepping down after medical leave, while Greg Brockman keeps leading product teams, and adds OpenAI’s go-to-market strategy to his role.

Top AI & SaaS Tools

  • Inkfluence (Life-time Deal): Turns a single prompt into a fully formatted, publish-ready book with chapters, cover design, and export options for PDF, EPUB, and DOCX

  • JustVibe: No-code search engine that builds custom interactive apps in your browser from your search queries [F-R-E-E]

  • Lispr: Mac and Windows app that lets you dictate or translate speech into any app with low latency and no audio stored [F-R-E-E]

  • Offer Max: Chrome extension that uses AI to simplify job applications, tailor documents, autofill forms, and prepare for interviews [F-R-E-E to Try]

  • Second Brain: AI memory system that connects information across multiple tools and helps AI remember projects, people, decisions, and preferences [F-R-E-E to Try]

Your employees are connecting AI to everything. Now what?

ChatGPT and Claude aren't just answering questions. Employees are connecting them directly to Notion, Linear, Jira, and the rest of your stack — with no security visibility into what data moves or what actions they take.

Harmonic Security gives your team the visibility to control it.

AI Tutorial

How to Build a Full-Stack CRM Using Google Antigravity

You do not need a large budget, a development team, or weeks of coding to build a custom CRM. With Google Antigravity and Firebase, you can turn a simple idea into a working, hosted business tool.

  1. Install Antigravity

    Download Google Antigravity, install the desktop app, and open it.

  2. Create Project Folder

    Create a folder for the CRM and open it inside Antigravity. A clean folder helps the agent stay focused on the project.

  3. Sign In With Google

    Make sure you are signed in to your Google account. Antigravity and Firebase will use this account during setup and deployment.

  4. Describe CRM

    Give Antigravity a prompt, such as:

    “Build ClientBase CRM with login, contacts, companies, deals, notes, search, and dashboard cards. Use React, Vite, TypeScript, Firebase Auth, Firestore, and Firebase Hosting. Create a plan before coding.”

  5. Review Plan

    Check the plan before allowing Antigravity to build. Remove unnecessary features such as:

    • Billing

    • Team workspaces

    • Advanced permissions

  6. Connect Firebase

    Let Antigravity build the app, then ask it to guide you through Firebase setup. Configure:

    • Firebase Authentication

    • Firestore

    • The Firebase project connection

  7. Deploy and Test

    Ask Antigravity to deploy the app using Firebase Hosting. Open the live URL and test login, contacts, deals, notes, search, and saved data.

Note: Refresh the app after saving records to make sure Firestore is storing the data correctly.

Top AI & Tech News

  • Samsung Health users are being asked to consent to their health data being used for AI training, or risk losing sync access and having their data deleted

  • Anthropic’s analysis of 1.2 million Claude Cowork sessions shows the tool is used mostly for routine office work, especially business process operations and content creation, rather than coding

  • Sam Altman has shifted from warning that AI could cause rapid job losses to saying he’s “pretty sure” it may create more jobs than it eliminates

  • Schwarz Group, owner of Lidl and Kaufland, wants to build one of the EU’s planned AI gigafactories in Germany as Europe pushes to cut dependence on US and Chinese tech

AI Art Spotlight

Create Your Own Neon Logo Like This Using the Prompt Below:

Create an ultra-detailed hyper-realistic 3D render of [logo], constructed from expertly bent neon tubing with a clean contemporary sign-making aesthetic, continuous luminous glass tubes following the exact logo shape, smooth curves, realistic tube thickness, bright internal glow, subtle colored bloom, delicate reflections on the glass surface, visible electrode detailing at connection points, refined mounting structure kept minimal, premium commercial signage look, isolated on a pure white seamless background, soft studio lighting balanced with the neon emission, gentle shadowing, macro detail, razor-sharp focus, photorealistic, 8k, 1:1, no watermark, no extra text

Model: ChatGPT Images 2.0 / Nano Banana 2

Prompt of the Day

99 Claude Secret Prompt Commands

Download it for a clearer view.

A practical collection of 99 Claude prompt commands designed to help you write more efficiently, think clearer, and work smarter. Covering email, writing, learning, planning, brainstorming, meetings, and content creation, these prompts give you ready to use ways to improve outputs, save time, and handle everyday tasks with less effort.

Featured AI Video

7 Claude AI Side Hustles That Pay $200+ a Day (Beginner-Friendly)