In partnership with

Welcome back!

OpenAI just gave ChatGPT a new superpower: Shopping Research, a built-in AI shopper that hunts for products, compares reviews, and tailors guides around your tastes. It’s OpenAI’s bid to become the front door of online buying. Let’s unpack…

Today’s Summary:

• 🛍️ ChatGPT becomes an AI shopping assistant

• 🔥 Anthropic slashes prices with Claude Opus 4.5

• 🤖 DeepMind unveils SIMA 2 embodied AI

• 🧩 Meta launches SAM 3 for segmentation

• 🦾 DeepMind hires ex-Boston Dynamics CTO

• 🔮 Sam Altman teases calm, screenless AI device

• 🛠️ 2 new tools

  TOP STORY

OpenAI turns ChatGPT into a shopping assistant

The Summary: OpenAI launched Shopping Research, a new feature in ChatGPT that acts as an interactive shopping assistant. It researches products across the web, asks clarifying questions, and builds personalized buyer’s guides based on user feedback and memory. The tool is rolling out to all logged-in users on mobile and web, with nearly unlimited use through the holidays.

Key details:

• Powered by GPT-5 fine-tuned with reinforcement learning

• 64% product accuracy compared to 37% for ChatGPT Search

• Users can tag items “Not interested” or “More like this”

• Pro users can get proactive shopping suggestions through Pulse

Why it matters: This tool allows ChatGPT to be a personal shopping agent that uses your personal data to make recommendations. The memory integration means the agent gets smarter about user preferences over time. For retailers, this creates a new gatekeeper in the buying journey.

FROM OUR PARTNERS

Boost Productivity with Smarter AI Prompts

Want to get the most out of ChatGPT?

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide

ANTHROPIC

Anthropic releases Claude Opus 4.5 with 3x price cut

The Summary: Anthropic’s new Claude Opus 4.5 lands with a major price drop and a focus on real-world software engineering. The model beats all previous records on the SWE-Bench Verified benchmark. With sharper reasoning, safer agent behavior, and deep integration into Chrome, Excel, and desktop apps, Anthropic positions Opus 4.5 as the practical workhorse for developers.

Key details:

• 3x cheaper API compared to Opus 4.1

• Uses 76% fewer tokens than Sonnet 4.5 for the same results

• Cost-per-task now rivals or beats Gemini 3 and GPT-5.1

• Claude Code now also runs in the desktop app

• Available for paid Claude tiers and API users

Why it matters: The release came within a week of big updates from OpenAI and Google, forcing prices down. For once, the market is doing what it should, with AI users winning from the intense competition.

System Card

FROM OUR PARTNERS

Teach Anything in Seconds with This Free AI Extension

From Boring to Brilliant: Training Videos Made Simple

Say goodbye to dense, static documents. And say hello to captivating how-to videos for your team using Guidde.

1️⃣ Create in Minutes: Simplify complex tasks into step-by-step guides using AI.
2️⃣ Real-Time Updates: Keep training content fresh and accurate with instant revisions.
3️⃣ Global Accessibility: Share guides in any language effortlessly.

Make training more impactful and inclusive today.

The best part? The browser extension is 100% free.

Check out Guidde

GOOGLE

DeepMind SIMA 2 learns, plays, and reasons in 3D worlds

The Summary: Google DeepMind has introduced SIMA 2, a Gemini-powered AI agent that can reason, plan, and learn inside 3D virtual environments. It understands goals, talks through its thinking, and improves through experience. SIMA 2 operates in unseen games and even inside AI-generated worlds. DeepMind calls it a major advance toward embodied, general-purpose AI.

Key details:

• Runs at near real-time speeds controlling mouse and screen input

• Approaches human-level reasoning on unseen environments

• Handles multimodal input: users can type, speak, sketch

• Implications for robotics: coupling SIMA 2 reasoning with control systems could connect virtual agents with robots

Why it matters: The breakthrough is a working prototype of embodied AI that learns transferable skills. By just watching pixels and pressing keys, it develops genuine understanding that generalizes. Most robotics research trains specialized models for narrow tasks with expensive human supervision. SIMA 2’s approach could accelerate robot development by pre-learning inside generated worlds.

QUICK NEWS

Quick news

• Meta launches SAM 3 for image and video segmentation

• DeepMind hires Boston Dynamics ex-CTO to lead robotics push

• Sam Altman teases OpenAI’s calm, screenless AI device

TOOLS

🥇 New tools

• Accent Guru - How strong is your accent when you speak English?

• Nao - Preview your data, run SQL queries, and fix your tables