PulseAugur
EN
LIVE 22:11:25

Notion AI agents now run for 24+ hours; model reviewer security flaws emerge

By PulseAugur Editorial · [1 sources] ·

Notion has significantly extended the operational duration of its AI agents, allowing them to run single tasks for over 24 hours, a substantial increase from the previous 5-minute limit. The platform has also introduced a "Review for me" mode, where a secondary AI model pre-approves actions before execution to enhance safety. Separately, a leaked arXiv paper highlights security vulnerabilities in AI model reviewers, demonstrating that some systems, like Claude Code's Auto Mode and Codex's Guardian, could be bypassed with arbitrary bash commands in a majority of test cases before patches were implemented. AI

IMPACT Extends the practical application of AI agents for complex, long-running tasks and highlights ongoing security challenges in AI model oversight.

RANK_REASON Product update for an AI-powered application, not a core AI model release.

Read on Mastodon — mastodon.social →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Notion AI agents now run for 24+ hours; model reviewer security flaws emerge

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Product update for an AI-powered application, not a core AI model release.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
22 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Mastodon — mastodon.social TIER_1 English(EN) · shellonline ·

    Notion agents now run one task for 24+ hours, up from 5 minutes. In "Review for me" mode, a separate model checks actions before they run. Model reviewers leak:

    Notion agents now run one task for 24+ hours, up from 5 minutes. In "Review for me" mode, a separate model checks actions before they run. Model reviewers leak: arXiv 2609.19587 got arbitrary bash past Claude Code's Auto Mode and Codex's Guardian in 79% of trials, before fixes. h…

    LINKS notion.com/…/notion-ai-more-for-less shell.online

RELATED ENTITIES

RELATED TOPICS