Loading story…
OpenAI launched Dots, persistent GPT-6 Astra-based agents hosted on OpenAI cloud computers, alongside ChatGPT Space, Pages, and collaborative slides. Dots can use browsers and more than 4,000 connected applications to continuously handle work such as monitoring feedback, updating sales materials, rerunning analyses, building and testing software changes, and preparing pull requests. Space provides a shared environment for employees, ChatGPT, and agents, while Pages and slides add AI-assisted document and presentation creation with real-time collaboration for Pro, Business, and Enterprise users.
OpenAI said Dots isolate agent activity from local user computers unless explicitly connected and distinguish read-only proactive research from actions that alter accounts, applications, or shared data. Higher-impact actions undergo automated review and may require user approval; additional safeguards include malicious-instruction defenses, safety monitoring that can pause or stop agents, and activity logs. For enterprises, OpenAI previewed specialist Dots using organization-provisioned identities, credentials, hardware, and controlled access, with planned Microsoft Agent 365 governance integration. The company also introduced Private Intelligence, an approval layer for sensitive enterprise data, including Zero Data Retention with Private Safety Processing, and said a future Private Inference preview will use confidential computing and verifiable controls.
Track how attackers are adapting to this technology.
3 events from the most recent confirmed update back to the earliest known activity.
OpenAI announced Private Intelligence, an enterprise data-access approval layer intended to protect sensitive content. It said Zero Data Retention with Private Safety Processing enables automated safety reviews without personnel accessing underlying content, and planned a Private Inference preview using confidential computing and verifiable controls.
At its fourth annual DevDay, OpenAI announced ChatGPT Space as a shared workspace for teams, ChatGPT, and Dots, along with the Pages collaborative document format and a new AI-assisted slides capability. Space and Pages were made available immediately, while slides were slated for release in the coming weeks.
OpenAI introduced Dots, persistent GPT-6 Astra-based agents that operate on OpenAI-hosted cloud computers, use browsers, and connect to more than 4,000 applications. The company described safeguards including restrictions on proactive research, review and approval controls for consequential actions, malicious-instruction protections, monitoring, and activity logs.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
Follow how adversaries are adapting to this technology, and where it touches your stack today.
3 references tracked. Mallory keeps watching after this page renders.
zdnet.com
Open sourcethenewstack.io
Open sourceopenai.com
Open sourceMap indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
OpenAI announced a broader rollout of its `GPT-5.6` model family and introduced **ChatGPT Work**, an enterprise-focused agentic AI service designed to automate multi-step tasks across applications, files, and business tools. The company said the new models—reported as variants including Sol, Terra, and Luna—deliver stronger benchmark performance while lowering inference costs and token consumption, and positioned them against rival Anthropic offerings on coding and agentic evaluations. OpenAI said ChatGPT Work can access a user’s desktop, browser, notes, and enterprise applications to carry out longer-running workflows such as creating spreadsheets, presentations, websites, and other business documents. The rollout begins on web and mobile for **Pro, Enterprise, and Edu** customers, with broader availability planned for other paid tiers, while OpenAI said `GPT-5.6` includes safety measures such as model protections, real-time checks, monitoring, and trust-based access controls; the wider release also follows earlier U.S. government restrictions tied to concerns about advanced cybersecurity and biology capabilities.
Jul 13, 2026
OpenAI announced **GPT-5.3-Codex**, positioning it as more than a code-generation model and expanding its availability across a new macOS desktop app plus CLI, IDE extension, and web interface, with API access described as forthcoming. OpenAI and third-party coverage said the model is intended to support broader software-lifecycle work (e.g., debugging, deployments, monitoring, tests, and documentation) and highlighted benchmark gains over prior Codex versions; reporting also pushed back on claims that “Codex built itself,” characterizing OpenAI’s statement as the model being *instrumental* in its own development rather than fully autonomous. In parallel, OpenAI unveiled **Frontier**, an enterprise framework aimed at building, deploying, and managing AI agents, including promised agent security features and an approach modeled on Palantir-style **forward-deployed engineers** working alongside customer teams. Separately, *MintMCP* launched an enterprise governance platform focused on deploying, monitoring, and securing AI agents and MCP servers, emphasizing audit trails, policy enforcement, observability/guardrails, and centralized access controls to reduce risks from privileged agents (e.g., credential exposure and data exfiltration). Other items in the set were not tied to these launches, including a general CIO column on SMB IT “quick fixes” creating long-term risk and a corporate internal AI learning program announcement from Hancom.
Jun 29, 2026
OpenAI introduced **Lockdown Mode** and **Elevated Risk** labels in *ChatGPT* to reduce exposure to **prompt injection** and related data-exfiltration risks when AI features interact with external systems. Lockdown Mode is positioned as an optional, advanced setting for higher-risk users and environments (notably *ChatGPT Enterprise*, *Edu*, *for Healthcare*, and *for Teachers*) that restricts tool access and limits how ChatGPT can reach outside systems; reported controls include disabling or constraining capabilities attackers could abuse via conversations or connected apps, and limiting browsing so that no live network requests leave OpenAI-controlled infrastructure (with browsing constrained to cached content). Admins can enable the setting via workspace controls and apply additional restrictions through dedicated roles, while Elevated Risk labels provide in-product warnings and guidance for features that increase risk when connecting to apps or the web, including across *ChatGPT*, *ChatGPT Atlas*, and *Codex*. Separate research highlighted how AI assistants with web-browsing/URL-fetching features can be abused as stealthy **command-and-control (C2) relays**, demonstrating a technique against **Microsoft Copilot** and **xAI Grok** that tunnels operator commands and victim data through legitimate AI web interfaces and can work without an API key or registered account. In parallel, the **European Parliament** reportedly disabled built-in AI tools on lawmakers’ work devices due to cybersecurity and privacy concerns about uploading sensitive correspondence to third-party cloud AI providers and uncertainty about what data is shared and retained. Other referenced material focused on general productivity customization of ChatGPT via “Custom Instructions,” rather than a specific security event or disclosure.
Jun 29, 2026