🚀 ChatGPT Ads Debut

✨ Welcome back, Entrepreneurs

This issue ads are entering ChatGPT, coding agents are speeding up real work, data stacks are becoming AI-native, creativity is being synthetic-rebuilt, and healthcare discovery is moving upstream before trials even begin.

💬 OpenAI begins testing ads in ChatGPT

OpenAI has started testing ads in ChatGPT for logged-in adult users on Free and Go plans in the U.S. Ads are clearly labeled, kept separate from responses, and do not influence answers. Conversations remain private, advertisers see only aggregated metrics, and paid tiers stay ad-free as OpenAI studies how ads can support broader access.

Highlights

• Ads appear only for Free and Go users; Plus, Pro, Business, and Education stay ad-free

• Answers remain independent and optimized for usefulness, not advertisers

• Advertisers never see chats, memory, or personal data—only aggregate views and clicks

• Users can dismiss ads, manage personalization, and delete ad data anytime

Why it matters:
This cautiously steps towards funding free AI access via ads, ensuring quality, privacy, and sustainable AI scaling while preserving user trust.

I built an AI app by talking to it — no code, no breakage

Most no-code tools fall apart the moment things get complex.
Appy.AI takes a different approach: living software that adapts in real time.
You build and edit AI agents just by chatting — for personal use, teams, or monetizable apps.
It feels less like using a tool and more like collaborating with an intelligent partner.

🔥Big AI moves

🔹Databricks tops $5.4B run-rate, accelerates AI-native data stack

Databricks reported a $5.4B revenue run-rate with 65%+ YoY growth and announced $7B+ in new investments, valuing the company at $134B. The funding will enhance Lakebase, its AI-focused serverless Postgres database, and Genie, its conversational AI assistant, advancing its operational AI and analytics.

🔹OpenAI rolls out GPT-5.3-Codex for advanced coding tasks

OpenAI has launched GPT-5.3-Codex, a new coding model that combines stronger reasoning with frontier-level software performance. Available in Cursor, GitHub Copilot, and VS Code, it scores 56.8% on SWE-Bench Pro, 77.3% on Terminal-Bench 2.0, runs 25% faster on agentic tasks, and supports long workflows.

🔹AI Deepfakes Orson Welles’s Lost “Ambersons”

Edward Saatchi's startup plans to use AI to recreate the missing 43 minutes of "The Magnificent Ambersons" by filming modern actors and overlaying the original cast's faces and voices. Rights and estate approval are pending, raising debate over whether this is historical restoration or a violation of authorship and performance consent.

The Hustle: Claude Hacks For Marketers

Some people use Claude to write emails. Others use it to basically run their entire business while they play Wordle.

This isn't just ChatGPT's cooler cousin. It's the AI that's quietly revolutionizing how smart people work – writing entire business plans, planning marketing campaigns, and basically becoming the intern you never have to pay.

The Hustle's new guide shows you exactly how the AI-literate are leaving everyone else behind. Subscribe for instant access.

Get our Claude playbook

🩺 AIHealthTech Insider: Aloe Targets Alzheimer’s

AI modeling flags an aloe vera compound that binds two enzymes linked to memory loss suggesting a plant-based candidate worth deeper lab and clinical validation. This issue explores how computation is accelerating drug discovery before symptoms or trials begin.

🎬 ByteDance launches Seedance 2.0, a multimodal AI video model with “reference control”

ByteDance's Seedance 2.0, in limited beta on Jimeng, creates 4–15 second clips from various media with automatic sound effects. The upgrade lets creators use reference videos for camera moves, motion, pacing, and effects, and swap characters or extend clips using “@asset” prompts.

Highlights

• Takes 4 modalities at once: image, video, audio, text

• Supports up to 12 total files (≤9 images, ≤3 videos, ≤3 audio; clips ≤15s)

• Copies camera language + complex motion from reference videos

• Can extend and stitch scenes into continuous shots (“keep filming”)

• Adds editing features like replace / remove / add characters in existing video

Why it matters:
This AI video tool lets you direct clips by controlling movement, speed, and style using real footage as a reference, streamlining short video creation. However, consistency and cost in real-world use remain uncertain.

⚡V0 by Vercel: Build Full Apps from a Single Prompt

V0 turns plain prompts into production-ready apps, no coding required.

How it works

• Describe what you want: landing page, dashboard, blog, anything

• V0 generates layouts, components, and UI instantly

• Refine the whole app with simple follow-up prompts

• Add integrations like OpenAI, Supabase, Clerk, Stripe

Pro tip: Batch-create branded prototypes fast and ideal for founders, designers, and devs shipping quickly.

⚙️ Sunday Special: Learning scales — credentials lag behind

This Sunday Special explores a shift in learning and creation: AI is now used for action, not just information. With 300 million weekly ChatGPT users learning by doing, degrees are becoming more important, and Anthropic is funding creators. It highlights how learning, hiring, and building are advancing faster than traditional institutions can keep up.