Agility Robotics, a Portland-based humanoid robotics startup that emerged from Oregon State University in 2015, announced plans to go public via SPAC merger at a $2.5 billion valuation. The company expects to raise approximately $620 million in gross proceeds to fund manufacturing scale-up and commercialization of its bipedal robots designed for warehouse and logistics work.
Enterprise organizations are now implementing spending caps and rate limits on AI usage after discovering that employees rapidly exhaust allocated budgets by using AI systems for low-value tasks like spell-checking, formatting, and other routine work. This shift from unlimited access to rationed consumption marks the end of an experimental phase.
Rep. Anna Paulina Luna (R-FL) clarified that her staff used AI to assist with formatting and editing in a defense bill amendment, though she explicitly denied using AI to draft the legislative language itself. The disclosure came after social media posts raised questions about AI's role in drafting federal legislation.
Anthropic released Claude Tag, a workplace AI agent that operates directly within Slack. The tool functions as a virtual team member: employees can tag it in conversations to summarize discussions, draft responses, analyze documents, or assist with tasks without leaving their chat interface.
Figma unveiled new AI-assisted features for motion graphics and shader generation at its Config 2026 conference, designed to help designers automate tedious animation work and experiment with visual effects more rapidly. The tools integrate directly into Figma's canvas, allowing designers to describe animations in natural language and have the AI generate the underlying code and visual output.
An immunologist named Derya Unutmaz used GPT-5 Pro to solve a three-year research puzzle about T cell behavior—work that had stalled despite conventional analysis. The AI model helped identify patterns and generate hypotheses that led to a breakthrough with implications for cancer treatment and autoimmune disease research.
Meta is piloting an AI companion app built into Facebook specifically for content creators. The tool integrates Meta's recently launched AI assistant and helps creators with tasks like ideation, caption generation, trend analysis, and content optimization. The beta test is underway with select creator accounts.
OpenAI and Broadcom have jointly developed Jalapeño, a specialized semiconductor designed to handle the computational demands of large language model inference—the process of running trained models to generate responses. The chip targets a critical bottleneck in AI deployment: the latency and power consumption of serving models to end users at scale.
OpenAI announced support for the Appia Foundation, a multi-stakeholder initiative developing shared evaluation frameworks, safety practices, and governance standards for advanced AI systems. The effort brings together AI developers, researchers, policymakers, and civil society to align on testing methodologies and risk management approaches.
Oracle announced 21,000 layoffs—roughly 10% of its workforce—as part of a strategic restructuring that frees up capital for aggressive investment in AI infrastructure. The company is funding a multi-billion-dollar expansion of data center capacity specifically designed to support AI workloads, partially through increased debt financing.
The White House issued an executive order dramatically shortening the timeline for government agencies and critical infrastructure operators to transition away from cryptographic systems vulnerable to future quantum computers. The order frames this as a national security imperative, warning that adversaries may already be capturing encrypted data today to decrypt when quantum capability matures.