OpenAI Launches Codex Security

In partnership with

OpenAI shipped both a code vulnerability hunter and a security acquisition. Anthropic mapped which jobs AI is really affecting and then sued the Pentagon to protect its business. Yann LeCun raised a billion dollars to bet against the LLM paradigm entirely. And NVIDIA set the stage for what could be the most important chip announcement of the year.

OpenAI Launches Codex Security, an AI Agent That Hunts and Patches Code Vulnerabilities

OpenAI released Codex Security in research preview on March 6, making it available free for the first month to ChatGPT Pro, Enterprise, Business, and Edu users.

The tool connects to a GitHub repository, builds a project-specific threat model in natural language, then searches for vulnerabilities and validates them inside sandboxed environments before surfacing findings ranked by real-world severity. During beta testing, false positive rates fell by more than 50 percent and over-reported severity findings dropped by over 90 percent.

The agent has already scanned more than 1.2 million commits, surfacing 792 critical findings and over 10,500 high-severity issues, and helped disclose 14 CVEs in major open-source projects including OpenSSH, GnuTLS, and Chromium. This positions OpenAI squarely in the fast-growing AI-enabled application security market alongside Anthropic, which launched a competing product weeks earlier. Source: (OpenAI)

2 Million Developers Use This Platform and You've Probably Never Heard of It

AI Agents Are Reading Your Docs. Are You Ready?

Last month, 48% of visitors to documentation sites across Mintlify were AI agents—not humans.

Claude Code, Cursor, and other coding agents are becoming the actual customers reading your docs. And they read everything.

This changes what good documentation means. Humans skim and forgive gaps. Agents methodically check every endpoint, read every guide, and compare you against alternatives with zero fatigue.

Your docs aren't just helping users anymore—they're your product's first interview with the machines deciding whether to recommend you.

That means:
→ Clear schema markup so agents can parse your content
→ Real benchmarks, not marketing fluff
→ Open endpoints agents can actually test
→ Honest comparisons that emphasize strengths without hype

In the agentic world, documentation becomes 10x more important. Companies that make their products machine-understandable will win distribution through AI.

Make Your Docs Agent-Ready

23 Ways to Profit with Nano Banana Pro

OpenAI Acquires AI Security Startup Promptfoo

OpenAI announced on March 9 that it is acquiring Promptfoo, an AI security startup founded in 2024 by Ian Webster and Michael D’Angelo. Promptfoo’s tools are used by more than 350,000 developers and over 25 percent of Fortune 500 companies to test AI systems for prompt injections, jailbreaks, data leaks, and other security risks during development.

Once the deal closes, Promptfoo’s technology will be integrated into OpenAI Frontier, the company’s enterprise platform for building and managing AI agents. Promptfoo will remain open source and will continue supporting multiple AI providers and models. The acquisition underscores the intensifying focus on agentic AI security as enterprises deploy autonomous agents into critical workflows. (Source: Openai)

System Prompt Guide

Best image Generator by Google in 2026

10 Claude Prompts to Do 4 Hours of Thinking in 40 Minutes

AI & Tech News

1.  Anthropic Publishes Landmark Study Mapping AI’s Real Impact on the Labor Market

Anthropic released a detailed economic study titled “Labor market impacts of AI: A new measure and early evidence,” introducing a framework called “observed exposure” that measures the gap between what AI can theoretically automate and what it is actually doing in workplaces today.

Using real usage data from its Claude model across roughly 800 U.S. occupations, the researchers found that computer programmers show 75 percent task coverage, the highest in the dataset, followed by customer service representatives and data entry workers. However, actual adoption remains a fraction of theoretical capability.

The paper found limited evidence that AI has materially affected unemployment so far, but flagged a 14 percent drop in the job-finding rate for young workers in AI-exposed occupations. The researchers warned that a scenario comparable to a “Great Recession for white-collar workers” is plausible as adoption deepens. Source: (Anthropic)

2.  Yann LeCun’s AMI Labs Raises $1.03 Billion to Build AI That Understands the Physical World

Turing Award winner Yann LeCun, who left Meta in late 2025, raised more than one billion dollars for AMI Labs at a $3.5 billion pre-money valuation, reportedly the largest seed round for a European AI startup. The Paris-based company is building “world models” that learn from physical reality rather than just from text, a direct challenge to the dominant large language model strategies pursued by OpenAI, Anthropic, and others. Investors include Cathay Innovation, Bezos Expeditions, HV Capital, Nvidia, and Temasek.

LeCun has been vocal in arguing that extending LLMs toward general intelligence is fundamentally limited, and that systems must be grounded in how the physical world works. AMI plans to open-source its technology and operate from offices in Paris, Montreal, Singapore, and New York. (Source: Techcrunch)

3.  Anthropic Sues the Pentagon Over Supply Chain Risk Designation

Anthropic filed a federal lawsuit against the U.S. Department of Defense to block a potential blacklisting after the Pentagon designated the company a “supply chain risk.” Court filings revealed significant financial exposure: Anthropic’s CFO warned the designation could reduce 2026 revenue by multiple billions of dollars, and that it has already disrupted deals worth hundreds of millions.

The company disclosed it has generated more than $5 billion in total revenue since launching commercial products, but has spent over $10 billion on training and deployment, remaining deeply unprofitable. The dispute arose after Anthropic attempted to set restrictions on how its AI could be used by the military. The case is being closely watched across the tech industry as a potential precedent for how national security authorities engage with private AI developers. (Source: Techcrunch)

4.  Anthropic Launches Code Review for Claude Code, Turning AI from Writer to Reviewer

Anthropic released Code Review on March 9, a multi-agent system built into Claude Code that automatically analyzes every pull request for logic errors, security risks, and subtle bugs that human reviewers routinely miss. When a pull request is opened, the system dispatches multiple AI agents that work in parallel, cross-verify each other’s findings, filter out false positives, and rank the remaining issues by severity. In Anthropic’s own internal usage, the share of pull requests receiving substantive review comments rose from 16 percent to 54 percent.

The feature is now available in research preview for Team and Enterprise customers, priced at $15 to $25 per review depending on complexity. Claude Code’s annualized run-rate revenue has surpassed $2.5 billion since launch, and the company reports that code output per engineer has grown 200 percent in the past year, making automated review an increasingly critical bottleneck to address. (Source: Claude)

5.  NVIDIA’s State of AI Report Shows 86% of Companies Plan to Increase AI Budgets in 2026

NVIDIA published its annual State of AI reports on March 10, drawing from more than 3,200 survey responses across financial services, retail, healthcare, telecommunications, and manufacturing. The core finding is unambiguous: 86 percent of respondents said their AI budgets will increase this year, with nearly 40 percent expecting increases of 10 percent or more.

The report found that 88 percent of companies said AI has helped increase annual revenue and 87 percent reported reduced costs. Agentic AI deployment has moved rapidly from experimentation to production, with telecommunications leading adoption at 48 percent, followed by retail at 47 percent. The data paints a picture of an industry in which AI is transitioning from a promising technology to an operational necessity across sectors, with measurable financial returns driving sustained investment. (Source: Nvidia)

6.  OpenAI Brings Interactive Math and Science Visuals to ChatGPT

OpenAI introduced dynamic visual explanations in ChatGPT on March 10, a feature that lets users interact with more than 70 core math and science concepts through manipulable variables and real-time visual feedback. Rather than presenting static diagrams, the tool allows learners to adjust inputs and immediately observe how formulas, graphs, and relationships respond.

For example, users exploring the Pythagorean theorem can change triangle side lengths and watch the hypotenuse update instantly. The feature builds on earlier education tools like study mode and QuizGPT, and it is available to all logged-in ChatGPT users worldwide. OpenAI cited research suggesting that interaction-based learning can lead to stronger conceptual understanding, noting that more than 140 million people already use ChatGPT weekly for help with math and science. (Source: OpenAI )

Top AI tools to check out…

I’ve been building a best AI tool directory where you’ll find the best of the best tool and their USECASES.

Thanks for your time!

Shailesh & OpenAILearning team