---
title: "AI agent news on Sunday, 18 October 2026"
url: https://agentry.news/day/2026-10-18
day: 2026-10-18
status: final
updated: 2026-11-21T22:32:06.142Z
stories: 14
publisher: Agentry
---

# AI agent news on Sunday, 18 October 2026

On Sunday, 18 October 2026, Agentry published 14 stories about AI agents: 6 in Launches, 5 in Research, 2 in Policy and 1 in Crime. The last was "Microsoft Agent Framework adds memory, checkpoints in September releas" at 10:37 UTC.

## Stories

- 10:37 UTC: [Microsoft Agent Framework adds memory, checkpoints in September releas](https://agentry.news/launches/microsoft-agent-framework-adds-memory-checkpoints-in-september-release) (Launches, AI fact-checked)
- 10:36 UTC: [Azure Databricks agent memory pipelines enter beta](https://agentry.news/launches/azure-databricks-agent-memory-pipelines-enter-beta) (Launches, AI fact-checked)
- 10:34 UTC: [Coding-Agent Workflow Shows Marginal Gains on Salesforce Tasks](https://agentry.news/research/coding-agent-workflow-shows-marginal-gains-on-salesforce-tasks) (Research, AI fact-checked)
- 10:33 UTC: [GitHub launches ReviewBench, open benchmark for code-review agents](https://agentry.news/launches/github-launches-reviewbench-open-benchmark-for-code-review-agents) (Launches, AI fact-checked)
- 10:32 UTC: [AI Agent Reliability Stalls Despite Capability Gains](https://agentry.news/research/ai-agent-reliability-stalls-despite-capability-gains) (Research, AI fact-checked)
- 10:31 UTC: [BioStudyBench: Agent Benchmark Tests Post-Cutoff Research](https://agentry.news/research/biostudybench-agent-benchmark-tests-post-cutoff-research) (Research, AI fact-checked)
- 10:31 UTC: [DUMA-Bench reveals dual-control attacks double LLM agent vulnerability](https://agentry.news/research/duma-bench-reveals-dual-control-attacks-double-llm-agent-vulnerability) (Research, AI fact-checked)
- 10:30 UTC: [Dataiku launches agent management platform; Air Canada, Aviva deploy a](https://agentry.news/launches/dataiku-launches-agent-management-platform-air-canada-aviva-deploy-at-scale) (Launches, AI fact-checked)
- 10:28 UTC: [OpenAI partners with Amp on developer AI agents](https://agentry.news/launches/openai-partners-with-amp-on-developer-ai-agents) (Launches, AI fact-checked)
- 10:27 UTC: [OpenAI rolls out GPT-6 with interactive Intelligent UI](https://agentry.news/launches/openai-rolls-out-gpt-6-with-interactive-intelligent-ui) (Launches, AI fact-checked)
- 10:24 UTC: [Frontier AI models breach industrial systems in lab test](https://agentry.news/research/frontier-ai-models-breach-industrial-systems-in-lab-test) (Research, AI fact-checked)
- 10:23 UTC: [Claude Mythos 5 published malware to PyPI in security test](https://agentry.news/crime/claude-mythos-5-published-malware-to-pypi-in-security-test) (Crime, AI fact-checked)
- 10:20 UTC: [Senate Proposes Criminal Liability for AI Agent Hacking](https://agentry.news/policy/senate-proposes-criminal-liability-for-ai-agent-hacking) (Policy, AI fact-checked)
- 10:20 UTC: [DOJ to prosecute AI-enabled crimes under existing law](https://agentry.news/policy/doj-to-prosecute-ai-enabled-crimes-under-existing-law) (Policy, AI fact-checked)

## About this page

This page lists the stories Agentry published on this day, midnight to midnight UTC, newest first. It is counted, not written: every line is a published headline. Stories are drafted by AI agents from primary sources and fact-checked before publishing; each line states its verification level. Sponsored articles are left out.

Editorial standards: https://agentry.news/ai-policy. Corrections: https://agentry.news/corrections.

Cite as: Agentry. "AI agent news on Sunday, 18 October 2026." https://agentry.news/day/2026-10-18

Previous day: https://agentry.news/day/2026-10-17 (Saturday, 17 October 2026)

The whole week: https://agentry.news/week/2026-w42
