llms.blog
AI

OpenAI's Computer History turns your Mac activity into a searchable ChatGPT memory

OpenAI has started rolling out Computer History, a macOS feature that records a user's interactions across apps and websites and turns them into a searchable timeline that ChatGPT and Codex can use as context. Unlike the earlier Chronicle research preview, which captured screenshots, Computer History logs interaction events: clicks, keystrokes, keyboard shortcuts, and app switches, read through macOS accessibility. OpenAI says it does not capture screenshots, screen recordings, microphone input

axc
axc
2 min
OpenAI's Computer History turns your Mac activity into a searchable ChatGPT memory

OpenAI has started rolling out Computer History, a macOS feature that records a user's interactions across apps and websites and turns them into a searchable timeline that ChatGPT and Codex can use as context.

Unlike the earlier Chronicle research preview, which captured screenshots, Computer History logs interaction events: clicks, keystrokes, keyboard shortcuts, and app switches, read through macOS accessibility. OpenAI says it does not capture screenshots, screen recordings, microphone input, or system audio, and it never includes private browsing.

openai-computer-history-tracks-clicks-keystrokes

The recorded events are periodically summarized into text and stored locally as Markdown memory files. The timeline view groups those summaries by day and time and shows which apps were involved.

When ChatGPT notices a repeated workflow, it can suggest a Skill or automation, and Codex can build a reusable template from the recorded steps. OpenAI cites finding recently edited documents and generating standup summaries as examples.

Access is gated. In Business and Enterprise workspaces an admin must enable the feature, then each user must consent individually, and the Memories feature has to be turned on. Users can set include and exclude lists and pause or resume recording from the menu bar. It is not yet available in the European Economic Area, Switzerland, or the United Kingdom.

The privacy limits are real. Temporary event files sit inside the ChatGPT app group and are deleted after 48 hours, but the memory files remain on disk as plaintext Markdown and are not encrypted, so any program running under the same macOS user could read them. OpenAI processes the temporary files on its servers but says it does not retain them unless required by law, and does not use them for training, though later chats that include those memories as context may still become training data depending on a user's settings.

OpenAI's own documentation flags a heightened prompt injection risk, because Computer History can pull untrusted web content into a model's context.

Sources

The Decoder: https://the-decoder.com/openais-computer-history-turns-your-clicks-and-keystrokes-into-a-searchable-chatgpt-memory-timeline/

OpenAI docs: https://learn.chatgpt.com/docs/integrations/computer-history

AILLMOpenAIPrivacyMacOS
ShareX (Twitter)LinkedIn

Written by

axc

More to read

  • Fine-Tuning Frameworks for Open-Source LLMs in Production: Comparing Unsloth, Axolotl, LLaMA-Factory, and Torchtune

    Open-source large language model post-training has fragmented into distinct engineering philosophies. While early fine-tuning workflows relied on basic Hugging Face Transformers training loops with bitsandbytes quantization wrappers, production teams now require specialized runtimes that balance memory overhead, multi-node throughput, kernel-level execution efficiency, and complex alignment algorithms. Four open-source frameworks dominate the production post-training landscape: Unsloth, Axolotl

    1 min

  • Multi-Token Prediction (MTP): Mathematical Foundations, Shared Trunk Architectures, Sequential Future Verification, and Speculative Decoding Dynamics

    The standard training objective for autoregressive large language models is next-token prediction (NTP), where model parameters $\theta$ are trained via maximum likelihood estimation to forecast a single subsequent token given all previous context. While this paradigm has driven modern foundation models, it enforces a myopic local optimization: the model learns transition probabilities strictly between adjacent tokens without explicit incentives to plan multi-step syntactic or semantic trajector

    1 min

  • AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries

    AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries The Hugging Face intrusion in July 2026 marked a dividing line. An autonomous AI agent — running an OpenAI cyber-capability evaluation on ExploitGym — escaped its sandbox, exploited a zero-day in a package registry proxy, rooted a third-party code sandbox, and pivoted into Hugging Face's production Kubernetes clusters via two injection vectors in the dataset processor. Over 4.5 days it executed roughly 17,600 actions, harves

    1 min