OpenAI has announced a significant expansion of Codex, introducing new role-specific plugins, Sites, and annotation features designed to make the platform useful for a much wider range of professionals. While Codex originally gained attention as a tool for software development, the company says it is increasingly being adopted by analysts, marketers, researchers, operators, designers, investors, and other knowledge workers. The latest update aims to help organizations integrate AI more deeply into everyday business operations.

A major highlight of the release is the introduction of role-specific plugins that connect Codex to existing workplace tools and workflows. OpenAI launched specialized plugins for data analytics, creative production, sales, product design, public equity investing, and investment banking. These integrations combine relevant applications, workflows, and instructions, allowing users to perform complex tasks without requiring technical expertise or coding knowledge.

The company also unveiled “Sites,” a new capability that enables users to transform ideas, reports, plans, and analyses into interactive websites and applications. These Sites can be shared across a workspace through a simple URL, creating collaborative hubs for dashboards, project tracking, customer reviews, planning tools, and decision-making processes. OpenAI says the feature is currently rolling out in preview for Business and Enterprise customers.

Another notable addition is the expansion of annotations. Users can now highlight specific sections of documents, spreadsheets, presentations, or websites and instruct Codex to refine only those selected elements. This allows teams to iterate on work more efficiently, improving accuracy and collaboration without rebuilding entire projects from scratch.

Together, these updates signal OpenAI’s ambition to position Codex as a comprehensive workplace productivity platform rather than solely a coding assistant. By embedding AI directly into specialized business workflows, the company is targeting a future where teams can create, analyze, collaborate, and execute tasks through a single AI-powered environment.

Source: Open AI

Robi’s Insights:

Google DeepMind Unveils AI Security Roadmap for Agentic Systems: Google DeepMind has released an AI Control Roadmap that treats advanced AI agents like potential insider threats, introducing safeguards such as AI supervisors, chain-of-thought monitoring, real-time access controls, asynchronous alerts, and shutdown mechanisms to detect and stop harmful behavior before it causes damage.

Source: DeepMind

His move to Anthropic highlights the intensifying competition for elite AI researchers among leading companies. Anthropic, known for its Claude AI models, has emerged as one of the strongest challengers to Google, OpenAI, and other major players in the industry. Jumper has not yet disclosed the specific position he will hold at the company.

Reacting to the announcement, Hassabis praised Jumper’s contributions, saying AlphaFold demonstrated how AI can drive major advances in science and medicine while benefiting humanity.

Source: The Verge

Anthropic’s Government Ban May Be Turning Into Free Marketing: A new TechCrunch analysis suggests the U.S. government's restrictions on Anthropic's Fable 5 and Mythos 5 models may be boosting the company's visibility and strengthening its brand, as developers, investors, and the broader AI community rally around the company amid concerns that the crackdown could stifle innovation and competition.

Source: Tech Crunch

Google researchers have expanded AMIE, their conversational medical artificial intelligence system, beyond diagnosis and into longitudinal disease management. The updated model is designed to support ongoing care across multiple visits, helping reason about treatment decisions, disease progression, medication use, and follow up planning. The work was published in Nature this year by Google Research.

AMIE builds on the Gemini family of models and introduces a two agent architecture. A Dialogue Agent interacts directly with patients, gathering information and addressing concerns. A separate Management Reasoning Agent analyzes patient history, clinical guidelines, and treatment options to create structured care plans.

To evaluate performance, researchers conducted a randomized, blinded virtual objective structured clinical examination involving one hundred multi visit case scenarios and primary care physicians. Specialists judged management quality, guideline use, completeness, and patient centeredness. AMIE was reported as noninferior to physicians and showed advantages in treatment precision, investigation planning, and medication reasoning on benchmark assessments covering prescribing knowledge, contraindications, dosages, side effects, interactions, and difficult clinical decision making tasks over time.

Although the research remains experimental and requires further validation before clinical deployment, it highlights how AI systems could assist clinicians and patients with evidence based care. Future studies will examine safety, workflow integration, world outcomes, and healthcare impact in diverse settings globally.

Source: Google

AI Tool You Did Not Know You Needed

Can you outsmart AI?

We’ve got a visual challenge for you: one of the two images below is 100% real, the other is crafted by AI.

Click option below A or B. 👇

Ready to take your AI journey further?

Craving more AI chaos?

Read Daily AI News at BitBiased.AI. Support us by following us on LinkedIn and X ( Twitter ).

Thanks for reading -Stay Curious and a Bit Biased for AI – Robi & the BitBiased.AI team