This website uses cookies

Read our Privacy policy and Terms of use for more information.

Hey everyone,

Would you let AI handle your next online purchase?

OpenAI is taking a step in that direction—with your approval before every payment. Meanwhile, Claude is joining video calls, and new models are arriving from Google, Anthropic, and Alibaba.

Here’s your weekly catch-up.

1. ChatGPT can help complete your purchases

OpenAI introduced ChatGPT Wallet, letting its agents pay on your behalf without seeing your actual card details.

You review the items, total, and selected card, then approve each purchase with a security check such as Face ID.

It’s initially rolling out through OpenAI’s dots agents to eligible US ChatGPT Pro subscribers aged 18 and older.

The takeaway: Your assistant could help you go from comparing products to checking out—with you approving the payment.

2. Claude gets faster—and cheaper to build with

Anthropic released Claude Fable 5.2 and Mythos 5.2, reporting improvements in coding, reasoning, and understanding images.

The company says they generate responses up to 30% faster than the previous versions, with lower standard API prices.

Fable is broadly available, while Mythos requires access through Anthropic’s verification programs.

The takeaway: People building with Claude could spend less and wait less on demanding tasks.

3. Google brings Gemini 4 to more people

Google introduced Gemini 4 Hydrogen, designed for everyday coding and tasks that require several steps.

It can work with large amounts of information at once and is rolling out across the Gemini app, Search, and developer tools. Free Gemini users can access it through automatic model selection when appropriate.

The takeaway: More users get access to Gemini 4 capabilities, with a focus on getting everyday work done quickly.

4. Claude joins the meeting

Anthropic introduced Claude Meetings in beta, starting with Google Meet on Team and Enterprise plans.

As a visible participant, Claude can answer questions, capture action items, and help draft documents during the conversation. It needs permission to use connected tools.

The takeaway: A team could leave a meeting with a working draft already started.

5. Alibaba launches Qwen4-Max

Alibaba released Qwen4-Max, its latest flagship AI model, built for tasks including coding, research, and understanding images.

It can process large amounts of information and is available through Alibaba’s hosted services. Smaller Qwen4 models with downloadable weights are planned for later.

The takeaway: Developers have another model to test against their own work—not just benchmark scores.

What would you hand to AI first: shopping, meeting notes, or coding?

Hit reply with your pick—I’d like to know why.

Found this useful? Forward it to a friend who wants to keep up with AI.

See you next week,
Activeprogrammer

Reply

Avatar

or to participate