AI Week #1

Welcome to the first AI Week, our short weekly round-up of AI news for freelancers. Grab a coffee: four stories, each in plain English, with a link to the original page so you can read more.

1. Google Cloud lets you put a hard cap on spending

On October 30, Google Cloud said spend caps in Cloud Budgets are now generally available. A cap can block pay-as-you-go use of eligible services once you hit your limit, and caps can now reset by themselves at the start of each month. Caps on AI services kick in within minutes, based on estimated live costs.

Why it matters for freelancers: if you test AI tools on a small cloud account, a cap can stop one runaway experiment from eating your monthly budget.

Source: Google Cloud Blog: spend caps now generally available

2. Microsoft shows off MAI Image 2.7 for photo editing

Microsoft AI announced MAI Image 2.7 on October 27. It is an image editing model, in preview, built for tricky changes like relighting a scene, changing text inside a picture, or restoring old images. You can try it on Arena, Artificial Analysis and MAI Playground. Microsoft says text-to-image creation is coming later.

Why it matters for freelancers: designers and social media managers may soon have another option for quick product shots and client edits. We haven’t tested it.

Source: Microsoft AI: MAI Image 2.7 debut

3. Claude Opus 5.5 and Sonnet 5.5 arrive in Google Antigravity

Google Cloud wrote on October 21 that Claude Opus 5.5 and Claude Sonnet 5.5 are now available in Google Antigravity for Google Cloud customers. An admin has to switch them on first. After that, developers can pick them in the model picker, next to Gemini models, across Antigravity 2.0, its CLI and its IDE extensions.

Why it matters for freelancers: if a client team uses Antigravity, you can mix models in one place, for example one model writes code and another checks it.

Source: Google Cloud Blog: Claude Opus 5.5 and Sonnet 5.5 in Antigravity

4. Gemini 4 Argon is Google’s new top model

Google introduced Gemini 4 Argon on October 13. It is aimed at long, complex work such as coding, finance and legal tasks. Developers can use it through the API, and Google AI Pro and Ultra subscribers get it in the Gemini app. The launch post lists an introductory API price of $2 per million input tokens and $10 per million output tokens.

Why it matters for freelancers: if you already pay for Google AI Pro, you can try a stronger model without adding another subscription.

Source: Google Blog: Gemini 4 Argon

Tools to try this week

• Best AI coding assistants: compare Antigravity, Cursor, GitHub Copilot and more.
• Best AI photo editors: our round-up of image editing tools for client work.
• Is ChatGPT Plus worth it?: a plain look at what a paid AI plan gives you.

Sources opened on November 1, 2026. Summaries are our own words; we have not tested the tools mentioned.