cd /news/artificial-intelligence/openai-introduces-computer-history-f… · home topics artificial-intelligence article
[ARTICLE · art-98841] src=cryptobriefing.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

OpenAI introduces Computer History feature in ChatGPT desktop app

OpenAI introduced Computer History, an opt-in feature for the ChatGPT desktop app on macOS, that records user activity across approved apps and websites to build a searchable timeline for ChatGPT and Codex. Announced on August 13, the feature replaces the earlier Chronicle screenshot-based approach and is initially available to Pro, Business, and Enterprise tier users, excluding those in the EEA, Switzerland, and the UK. Ari Weinstein, Product and Engineering manager at OpenAI, said the feature automatically ignores incognito or private browsing content.

read3 min views1 publishedAug 16, 2026
OpenAI introduces Computer History feature in ChatGPT desktop app
Image: Cryptobriefing (auto-discovered)

Via gizmodo.com

The new opt-in macOS tool tracks your desktop activity to build a searchable timeline that ChatGPT and Codex can reference, replacing the older screenshot-based Chronicle feature.

OpenAI just gave ChatGPT a photographic memory for everything you do on your Mac. The company’s new Computer History feature, announced on August 13, records user activity across approved apps and websites on macOS, building a searchable timeline that both ChatGPT and Codex can tap into when you ask for help.

How Computer History works #

The feature tracks actions like typing, clicks, and keyboard shortcuts through macOS accessibility APIs. That raw activity data gets converted into structured memories and a timeline that ChatGPT and Codex can reference to provide context-aware assistance.

The practical upside is significant. If you abandon a half-finished task, ChatGPT can pick up where you left off. It can suggest automations based on patterns it notices in your workflow. And when you make a request, it doesn’t start from zero, it already has context about what you’ve been working on.

Computer History replaces Chronicle, an earlier research preview that relied on periodic screen captures to achieve a similar goal. The new approach is less intrusive by design, pulling structured interaction data rather than snapping screenshots of your display.

Ari Weinstein, Product and Engineering manager at OpenAI, noted on X that Computer History will automatically ignore content viewed in incognito or private browser tabs. Weinstein, who previously worked at Apple, led the team that developed this feature as an evolution of the Memories system OpenAI first introduced in 2024.

Privacy controls and availability #

Computer History is opt-in and off by default. Users have to actively choose to turn it on, a meaningful distinction from the opt-out model that has tripped up other companies rolling out ambient data collection.

Beyond the initial toggle, the controls are fairly granular. Users can data collection at any time, exclude specific apps and websites from being tracked, and delete individual entries from the timeline. For organizations on team plans, administrators get their own set of access controls to manage how the feature operates across their workforce.

The initial rollout is limited to Pro, Business, and Enterprise tier users on macOS. Users in the European Economic Area, Switzerland, and the United Kingdom won’t have access at launch, a familiar pattern for AI features that bump up against the EU’s stricter data protection regime under GDPR.

The bigger picture for desktop AI #

Microsoft has been chasing this same idea with its Recall feature for Windows, which also captures desktop activity to create a searchable timeline. That product faced intense privacy backlash when announced in 2024, leading to delays and redesigns. OpenAI seems to have learned from Microsoft’s stumble by making Computer History opt-in from day one and by using accessibility APIs rather than screen captures.

The shift from Chronicle’s screenshot approach to structured interaction data is worth noting on its own. Screenshots are blunt instruments. They capture everything visible on screen, including sensitive information that happens to be in view. Tracking discrete interactions like clicks and keystrokes is more surgical, though it raises its own questions about keystroke logging and what data gets stored, where, and for how long.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our

Editorial Policy.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @openai 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/openai-introduces-co…] indexed:0 read:3min 2026-08-16 ·