Descriptions:
Matt Wolfe’s weekly AI news roundup for the week of February 27, 2026 opens with Google DeepMind’s Nano Banana 2 image generation model, now available for free in Gemini, AI Studio, and Vertex. The model features real-time web search grounding for more accurate subject rendering, strong text fidelity, and roughly half the generation time of its predecessor—demonstrated live with enamel pin creation and AI-generated infographics.
A significant portion of the video covers Perplexity’s newly launched Computer agent, which routes tasks across 19 different AI models and integrates with tools including Slack, Airtable, Discord, and ClickUp. Wolfe contrasts it with OpenClaw: Perplexity Computer is a managed, turnkey cloud experience, while OpenClaw prioritizes local control and data ownership. The catch is a steep entry price—currently limited to Perplexity’s $200/month Max plan.
The episode’s most consequential segment revisits the escalating standoff between Anthropic and the US Department of Defense. Defense Secretary Pete Hegseth set a Friday deadline for Anthropic to drop its two usage restrictions—prohibiting mass surveillance of US citizens and fully autonomous weapons with no human in the loop—threatening to designate Anthropic as a supply chain risk if it refused. That designation, normally reserved for foreign adversaries, would effectively blacklist Anthropic from the entire defense contracting ecosystem. Wolfe frames this as one of the most important ongoing stories in AI, with global implications for how safety policies interact with national security requirements.
📺 Source: Matt Wolfe · Published February 27, 2026
🏷️ Format: Roundup
