Hello AI Enthusiast,
This week, Anthropic found a clever way to make Claude more visual without building an image model, while Nvidia is trying to make AI agents safe enough for the corporate world. Meanwhile, brands are quietly rewriting their review strategies, because in the age of AI search, your star ratings may matter more than your ad budget. Let's get into it.
The Big Picture 🔊
Claude Now Builds Inline Visuals During Conversations
Anthropic has added interactive charts, diagrams, and visualizations directly into Claude's chat responses, available to all users. Unlike artifacts, these visuals appear inline rather than in a side panel. Claude decides when a visual would help, or users can request one directly and ask for adjustments as the conversation develops.
|
||
|
Anthropic skipped image generation entirely and built visuals using SVG, a format AI can generate through code alone. It is a clever workaround, not a new model. While competitors train separate models to generate images, Anthropic just got better at rendering what it already does best: writing code. |
Nvidia Launches NemoClaw to Make AI Agents Safer for Enterprises
Nvidia has announced NemoClaw, an open-source stack that adds privacy and security controls to OpenClaw, the fastest-growing open-source AI agent platform in history. NemoClaw installs with a single command and uses OpenShell, a sandboxing runtime that limits agents' access to sensitive data and enforces policy-based security guardrails.
|
||
|
NemoClaw addresses some known security problems with OpenClaw, like restricting file access and blocking unauthorized connections. But the underlying issue remains: AI agents can do a lot, and defining what they cannot do is not the same as making them safe. The hard problems are still unsolved. |
Companies are racing to deploy AI agents, but most teams are still figuring out the basics. Before you scale, make sure your people know what they are actually working with. Our Corporate AI Training helps teams cut through the hype and use AI where it genuinely adds value.
Customer Reviews Are the New SEO in the Age of AI Search
AI platforms like ChatGPT and Perplexity are reshaping product discovery, using customer reviews as a key input for recommendations. ChatGPT alone receives over 84 million shopping-related queries weekly in the US. Brands are responding by actively collecting more reviews across public platforms like Reddit and Yelp, offering incentives to customers, and treating reviews as a core part of their visibility strategy.
|
||
|
Brands are treating reviews the way they once treated SEO, optimizing for a system they do not fully control. The real risk is that as more companies game the process with incentives and volume, AI recommendations become less trustworthy. And without verified human data behind those reviews, the signal quickly turns to noise. |
AI Academy Insider: "My first inclination would be to create an agent that would spam reviews, which is why platforms will eventually need verified human data to separate real feedback from manufactured noise.”
Bits and Bobs 🗞️
OpenAI's new GPT-5.4 mini and nano models offer faster, more efficient AI performance for coding, multimodal tasks, and high-volume workflows at lower costs.
Microsoft's new Copilot Health uses AI to combine your medical data and wearable info into clear, personalized health insights that help you better understand your body and prepare for doctor visits.
Google Maps' new Ask Maps uses Gemini AI to answer complex location questions conversationally, while Immersive Navigation provides a 3D, AI-powered driving experience with detailed real-world visuals and guidance.
An autonomous AI agent discovered a critical unprotected SQL injection vulnerability in McKinsey's internal AI platform, highlighting that AI prompt layers are a new, often overlooked security risk.
Gamma Imagine uses AI and text prompts to help users easily create brand-specific visual assets like charts and infographics without needing advanced design skills.
Claude Opus 4.6 and Sonnet 4.6 now offer a full 1 million token context window at standard pricing, enabling models to handle and recall extensive information without extra cost.
Mistral Small 4 is an open-source AI model that combines chat, reasoning, coding, and multimodal abilities into one efficient, customizable tool.
Encyclopedia Britannica and Merriam-Webster sued OpenAI, accusing it of using their copyrighted articles without permission to train its AI models and generate competing content.
From Our Founder’s Channels 🤳
In his latest article, our founder Gianluca Mauro gives an honest take on where AI is taking us and what the future of work will look like. Read it here
LOLgorithms 😂
Anthropic is cooking.
That's a wrap on our newsletter! Before you go, here’s a quick recap of our offerings:
AI Academy Membership: Get 12 months of access to all our cohort-based programs and on-demand courses.
AI Agent Bootcamp: Accelerate processes and solve business problems by building AI Agents, without coding.
Practical Introduction to ChatGPT: A free course on using ChatGPT confidently, understanding its workings, and exploring its potential.
Customized Corporate Training: Equip your team with the skills they need to unlock the potential of AI in your business.
Catch you next week! 👋
