{"generated_at":"2026-06-15T15:11:20.746371+00:00","window_days":21,"storylines":[{"slug":"deep-research","label":"Deep Research","item_count":3,"day_count":3,"source_count":3,"first_seen":"2026-05-27T07:45:00+00:00","last_updated":"2026-06-15T13:56:33+00:00","latest_title":"Build context-rich research agents with Deep Agents and Bedrock AgentCore","days":["2026-05-27","2026-06-08","2026-06-15"],"member_sids":["db3f935ead18d952","c5cd09a4f478ae3e","2f2101adc8737656"],"member_urls":["https://www.infoq.com/news/2026/05/kulkarni-deep-research-agents/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=AI%2C+ML+%26+Data+Engineering","http://arxiv.org/abs/2606.09748v1","https://aws.amazon.com/blogs/machine-learning/build-context-rich-research-agents-with-deep-agents-and-bedrock-agentcore"],"editorial":{"tldr":"Deep research agents — systems that run multi-step research using dynamic reasoning and multi-hop retrieval — moved from field notes toward repeatable practice over three weeks. InfoQ opened with Sarang Kulkarni's production lessons on building them; an arXiv paper then flagged an evaluation gap, arguing existing benchmarks only score single-shot output and proposing multi-turn evaluation under process-level feedback; AWS closed with an end-to-end build using Deep Agents and Bedrock AgentCore.","stale":false}},{"slug":"claude-fable","label":"Fable Mythos","item_count":7,"day_count":5,"source_count":4,"first_seen":"2026-06-10T03:50:21+00:00","last_updated":"2026-06-15T05:01:00+00:00","via_scout":true,"latest_title":"Anthropic Releases and Temporarily Suspends Claude Fable 5","days":["2026-06-10","2026-06-11","2026-06-12","2026-06-13","2026-06-15"],"member_sids":["d232148f3de4820c","c9ab8f8aa14fe295","001317287cdd9d97","e11fcea9cd94bb20","e464b7a63a6a0397","1a3926308e38dbe8","575139a8c2fbbeb4"],"member_urls":["https://www.latent.space/p/ainews-anthropic-claude-fable-5-mythos","https://simonwillison.net/2026/Jun/11/fable-is-relentlessly-proactive/#atom-everything","https://www.anthropic.com/news/fable-mythos-access","https://www.anthropic.com/news/claude-fable-5-mythos-5","https://simonwillison.net/2026/Jun/13/us-government-directive-to-suspend-access/#atom-everything","https://www.latent.space/p/ainews-fable-and-mythos-officially","https://www.infoq.com/news/2026/06/claude-5-release/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=AI%2C+ML+%26+Data+Engineering"],"editorial":{"tldr":"Anthropic launched Claude Fable 5 and the higher Mythos 5 tier in mid-June, billing the Mythos-class model as made safe for general use, though launch coverage immediately flagged controversial usage terms; an early hands-on found Fable 5 \"relentlessly proactive.\" Anthropic published access details for the two tiers. Then on Jun 13 the story turned: a US government export-control directive, citing national security, suspended access to Fable 5 and Mythos 5, and industry coverage reframed the launch as models deemed too dangerous to release. A later recap framed the takedown as a temporary suspension.","stale":false}},{"slug":"dynamic-workflows","label":"Dynamic Workflows","item_count":3,"day_count":3,"source_count":3,"first_seen":"2026-05-29T02:07:24+00:00","last_updated":"2026-06-02T00:00:00+00:00","latest_title":"A Harness For Every Task Dynamic Workflows In Claude Code","days":["2026-05-29","2026-06-01","2026-06-02"],"member_sids":["008b880e906f9835","72c72052beef8f43","bbb7a8dae33b3ae4"],"member_urls":["https://www.latent.space/p/ainews-anthropic-raises-965b-series","https://www.infoq.com/news/2026/06/dynamic-workflows-claude-code/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=AI%2C+ML+%26+Data+Engineering","https://claude.com/blog/a-harness-for-every-task-dynamic-workflows-in-claude-code"],"editorial":{"tldr":"Anthropic introduced Dynamic Workflows (“ultracode”) for Claude Code, first surfaced in late May alongside the Opus 4.8 release and detailed over the following days. The capability coordinates large numbers of agents within a single workflow to take on complex engineering tasks, with Anthropic's own “A Harness For Every Task” post laying out the model.","stale":false}}]}