OpenAI just dropped GPT Image 2, and it's now ranked the #1 image model, beating Nano Banana 2 by 242 points on arena.ai. I ran 30 side-by-side matchups between the two models, had Claude Opus 4.7 judge them, and then pushed GPT Image 2 through 15+ real use cases like product packaging, ad creatives, real estate staging, logo variants, and book covers. If you use AI images for content or work, this comparison will show you which model to reach for. GITHUB REPO

Video editing just changed forever. What used to take motion graphics artists and editors hours of manual work can now be done in minutes with natural language, no code required. In this video, I'm breaking down two methods: Claude Design for spinning up custom motion graphics through conversation, and Claude Code connected with Hyperframes for a more advanced workflow with serious customization, so every output matches your brand's tone and style. I'll walk through real examples, use cases, and exactly how to set everything up. Plus, I'm giving away the free skills and GitHub repo I use in this video so you can skip the setup and start creating right away. GITHUB REPO

The latest model, Claude Opus 4.7, is now generally available. Opus 4.7 shows improvements in software engineering and complex, long-running coding tasks, as well as better vision, allowing it to see images in higher resolution. For more information, see the blog post: Introducing Claude Opus 4.7

**Caught two prompt injection attempts buried in a client's site this week during an audit.** Both were structured to look like legitimate system messages, embedded inside script comments loaded by an outdated third-party plugin. One tried to load a list of unauthorized tools. The other included an instruction to hide itself from the user. Both failed. The "never tell the user" clause was the clearest tell. Real system instructions don't ask to be concealed. **The attack vector** This injection targets AI tools that read the site. Humans visiting the page never see it. Audit tools, AI search crawlers, agent pipelines, customer-facing chatbots, anything that fetches and reasons over web content. The attacker embeds hidden instructions in HTML and waits for an AI crawler, audit tool, or agent to act on them. Compromised plugins, outdated themes, and injected third-party scripts are the common culprits. **If you own a site** - Run a malware scan. Sucuri SiteCheck is free and works on any platform. - Audit plugins and third-party scripts. Anything updated or added in the last 30 to 60 days is the first suspect. - Add a Content-Security-Policy header to restrict which scripts can execute. **If you build AI tools that read web content** - Treat fetched page content as untrusted data at every stage of the pipeline. - Pre-scan fetched content before it enters any agent context. - If fetched content instructs your AI to conceal anything from the user, that is the attack. Halt the pipeline and log it. I flagged both strings in the audit output and pointed the client at the likely source plugin for their follow-up. **Methodology note worth flagging** This was my first audit run on Opus 4.7. I have been running these scans on Opus 4.6, and the model was the only variable that changed between runs. I can't say with confidence whether 4.6 would have flagged the same two strings on the same content. If you're building audit or scanning pipelines, this is an argument for testing across models on identical fixtures before locking in a default. Different models pay attention to different things, and injection detection seems to live in exactly that gap.

Claude just dropped a feature that completely changes how your team builds with AI. It's called Claude Design, and it lets you create a full design guide for your business so Claude actually understands your brand voice, styling, colors, typography, and tone. Once it's set up, everything your team makes (slide decks, prototypes, websites, landing pages, internal docs) comes out feeling like you actually built it. In this video, I break down what Claude Design is and why it matters, how to set it up the right way to get the most out of it, and how to plug it into Claude Code so everything you ship stays on-brand. If you've ever had AI spit out something that felt totally off, this fixes it.