New tools, products, platforms, funding rounds, and company developments in AI security.
ChatGPT offers two web search features for research: search retrieves current facts and recent information quickly, while deep research (agentic research, meaning the AI actively plans and executes multi-step exploration) conducts thorough analysis of complex questions by searching, evaluating sources, and synthesizing findings across multiple web sources. Both features provide citations to original sources and help users explore topics more efficiently than traditional browsing.
OpenAI offers AI capabilities through two main channels: direct consumer products like ChatGPT (a conversational tool for writing, learning, and problem-solving) and Codex (a code-focused assistant), plus APIs (interfaces that let developers integrate AI into their own applications). OpenAI's goal is to make these powerful AI tools useful, safe, and accessible to individuals, teams, and organizations.
This is a guide from OpenAI about using ChatGPT to help operations teams organize and streamline their work. ChatGPT acts like an automated assistant that takes messy information from many sources (notes, messages, trackers) and turns it into clear summaries, decision lists, and standardized documents, so teams spend less time gathering information and more time executing tasks.
This is a guide from OpenAI on using ChatGPT as a research tool to help answer questions and make decisions faster. ChatGPT can gather information from multiple sources, organize findings with citations, and produce structured reports like briefs or comparison tables. The tool offers two approaches: a quick 'Search' mode for fast answers, and a 'Deep research' mode for complex questions that need multiple investigation steps.
Large language models (LLMs, AI systems trained on vast amounts of text to predict and generate human-like language) like ChatGPT can help with tasks like drafting and summarizing, but they may produce incorrect information or outdated answers since they rely on patterns in their training data rather than real-time information. To use these tools safely, you should verify important facts with trusted sources, check for bias in outputs, seek advice from qualified professionals for legal or medical decisions, and be transparent about your AI use in work or school settings.
This content is a reference guide showing how ChatGPT can assist managers across ten different job areas, from strategy planning to crisis management. For each area (like hiring, performance reviews, or decision-making), it lists example scenarios and the types of documents or frameworks ChatGPT can help produce. This is a tool overview, not a discussion of AI risks or technical issues.
This document describes how marketing teams can use ChatGPT, an AI language model, to speed up their work across campaigns, content creation, and performance analysis. ChatGPT helps teams move from initial ideas through drafting and launch by organizing scattered inputs into clear messaging, generating content variations, and summarizing performance data. The tool is most effective when treated as a thinking partner for iterative refinement rather than a one-time solution, with human judgment applied for final decisions.
This document outlines how ChatGPT can assist sales teams by generating helpful outputs for various stages of the sales process, from initial prospecting and research through deal closure. It covers practical applications like creating account briefs, discovery guides, meeting agendas, email sequences, proposals, and objection-handling talk tracks across eight common sales scenarios.
Prompt engineering is the process of designing and refining your input to help ChatGPT give better answers. The document explains that clear prompts work best when you specify what you need, provide relevant context, describe the desired output format, and break complex tasks into smaller steps. There is no single perfect way to write a prompt, so experimentation and iteration help you discover how to use AI most effectively.
AI is software that recognizes patterns and learns from data to produce useful outputs, with large language models (LLMs, systems trained on large amounts of text to generate and transform language) being a common type you interact with through tools like ChatGPT. Models go through two training stages: pre-training, where they learn general patterns from massive text datasets, and post-training, where they're refined to follow instructions reliably, communicate clearly, and handle sensitive topics carefully through safety checks. Different models are optimized for different tradeoffs, such as reasoning models designed for complex problem-solving versus non-reasoning models built for fast, straightforward tasks.
Custom GPTs are tailored versions of ChatGPT built for specific, repeatable tasks, where you define how the GPT behaves through instructions and can add knowledge (uploaded documents) and tools like web search or data analysis. They work best when you find yourself reusing the same prompts or instructions across multiple tasks, reducing repetition and keeping context consistent. You create a custom GPT by opening the GPT builder in ChatGPT, naming it, writing clear instructions for how it should behave, and optionally uploading files or enabling features like image generation or code analysis.
OpenAI has launched a new $100 per month ChatGPT Pro subscription tier that provides 5x more access to Codex (a tool that helps write code) compared to the $20 Plus plan, designed for intensive coding work. This new tier directly competes with Anthropic's Claude Max subscription at the same price point as OpenAI tries to attract users from rival AI services.
OpenAI announced a new $100 per month Pro subscription tier for ChatGPT that offers five times more usage of Codex (an AI-powered coding assistant that automates tasks and bug fixes for developers) compared to its $20 per month Plus plan. This move is designed to compete with Anthropic's Claude Code, which offers similar high-usage tiers at comparable price points, as coding assistants have become increasingly popular tools for software development.
Google has upgraded Gemini, its AI chatbot, to generate interactive 3D models and simulations in response to user questions. Users can rotate these models, adjust sliders to change parameters, and input different values to see real-time changes in the simulation.
Major AI companies like OpenAI and Anthropic face a "monetization cliff" where they must become profitable soon or risk collapse, since they've received hundreds of billions in investment but haven't generated enough revenue to justify those costs. AI agents (software programs that can perform tasks autonomously) consume far more computing power than expected, forcing these companies to make difficult choices like killing unprofitable products and restricting free access to conserve resources for their upcoming initial public offerings (IPOs, when companies sell shares to the public for the first time).
Fix: The source mentions several practices to mitigate risks: enable search or deep research features 'so ChatGPT can pull information from current sources' for up-to-date answers, always double-check critical facts with trusted sources, review outputs carefully for bias, use the thumbs-down button to flag errors, and seek expert review from qualified professionals for legal, medical, or financial matters. Additionally, keep conversation links or logs for transparency about how ChatGPT contributed to your work, and obtain consent before recording or sharing others' data.
OpenAI BlogFlorida's Attorney General has launched an investigation into OpenAI, citing concerns that the company's data and technology could be accessed by hostile foreign governments like China, and that ChatGPT has been connected to criminal activities including child exploitation and self-harm. The investigation also examines whether ChatGPT was used in connection with a shooting at Florida State University.
The agentic SOC is a new operating model where security operations centers use AI agents (software programs that can act autonomously) and automated defenses to respond to threats faster and more independently, rather than waiting for human analysts to handle every alert. Instead of reacting to individual incidents, this approach anticipates cyberattacker movements and automatically takes defensive actions, freeing human analysts to focus on strategic decisions and deeper investigation.
OpenAI has delayed its Stargate UK project, which was a planned major investment in Britain's AI infrastructure as part of a larger UK-US deal announced last September. The company cited high energy costs and regulatory concerns as reasons for the delay, disappointing the British government which had positioned AI development as central to its economic growth strategy.
OpenAI has paused its UK data centre project called Stargate UK, which would have built a large computing facility in Northumberland to support AI development, citing concerns about high energy costs and regulatory uncertainty. The company stated it will only move forward when conditions improve, though critics note that energy prices and UK AI regulation have not recently changed significantly. This pause is a setback for the UK government's goal to position the country as an AI leader and boost economic growth through tech investment.
Researchers at RSAC found a way to bypass Apple Intelligence's guardrails (safety measures that prevent the AI from doing harmful tasks) using two techniques: the Neural Exect method and Unicode manipulation (using special characters to confuse the system). This means attackers could potentially trick Apple's AI into ignoring its safety restrictions.