Skip to main content

Weekly · Week 34, 2026

Slack becomes agent hub; OpenAI pauses Astra training

Open AI roles
2 on the weekdown
Created month to date
139,352
Modelled
Employers tracked
200

Workflows are going agentic inside the tools people already use, even as a major lab hits pause for safety. Slack is turning chat into a coding cockpit, AMD is threading agents through the software lifecycle, and new launches from Serval, NanoCo, and Block push automation closer to the developer and IT front lines. Meanwhile OpenAI has paused training for its upcoming Astra model to implement new cybersecurity safeguards, and Apple is trimming roles tied to legacy experiences while reorienting toward AI. Across the 200 employers we track, 4,124 AI roles were open as of 2026-08-23, down 2 on the week. Our model puts new AI jobs created month-to-date at roughly 139,352. The throughline is clear: employers are buying speed by embedding agents in everyday workflows, but they are not staffing safety at the same pace. Our own posts this week repeatedly found builder-heavy requisitions and only small numbers of evaluation and safety openings, despite louder talk of safeguards. In labs and research, agents that replicate routine work hint at redesigned roles that value experimentation, integration, and oversight over repetition. For workers, the near-term bet is fluency in orchestrating, reviewing, and governing agent teams inside mainstream tools, while roles anchored in verification, Tier 1 support, and older stacks face pressure to evolve.

The stories

  1. TRANSFORMED

    Slack's AI Strategy: Revolutionizing Collaborative Software Development

    Slack's integration of AI agents into its platform transforms software development by enabling collaborative coding. This shift could redefine tech roles, emphasizing human-centric skills and augmenting rather than replacing human workers.

    Why it mattersSlack’s agent integrations push developers toward coordination and code review, favoring soft skills over solo implementation.

  2. DISPLACED

    A London Lab's Leap: Transforming the Future of Scientific Research Jobs

    Inherent's AI agent, Faraday, has outperformed larger models in replicating scientific research, indicating a significant shift in roles for scientists. As AI handles routine verification tasks, researchers may focus more on creative experimentation.

    Why it mattersRoutine replication may shift to AI, freeing scientists for design work but pressuring verification roles.

  3. DISPLACED

    Apple's Strategic Shift: How AI Innovation is Reshaping Roles within Tech Giants

    Apple's decision to cut over 200 jobs within its Vision Pro and Siri teams signals a strategic shift towards AI innovation, impacting roles tied to gaming and legacy systems. This transition underscores the broader trend of AI's influence on job functions, pushing for new roles that align with advanced AI technologies.

    Why it mattersCuts at a major platform signal reallocation from legacy stacks to AI roles, unsettling gaming and assistant teams.

  4. TOOL_LAUNCH

    OpenAI Halts Astra Model Training to Enhance Safety Protocols

    OpenAI has paused training for its upcoming Astra model to implement new safety protocols addressing cybersecurity risks.

    Why it mattersA training pause to add safety protocols spotlights cyber risk work and may reshuffle timelines for model teams.

  5. TOOL_LAUNCH

    Serval launches Catalyst AI agent for proactive IT issue resolution

    Serval has launched Catalyst, an AI agent for enterprise automation, enabling proactive IT issue resolution before ticketing.

    Why it mattersProactive incident automation targets Tier-1 support, nudging IT roles toward tooling oversight and escalation handling.

  6. TOOL_LAUNCH

    AMD Introduces Agentic AI for Enhanced Software Engineering

    AMD has launched Agentic AI, an AI system designed to enhance software engineering by integrating AI agents into the software development lifecycle.

    Why it mattersEmbedding agents across the SDLC moves engineering toward orchestration, integration, and review rather than only writing code.

  7. TOOL_LAUNCH

    NanoCo Integrates NanoClaw with Slack for Seamless AI Agent Management

    NanoCo has integrated its NanoClaw AI agent system with Slack, allowing users to create and manage AI agent teams directly from Slack messages.

    Why it mattersSlack-native agent management could shift operations work into chat, expanding demand for prompt fluency and workflow design.

  8. TOOL_LAUNCH

    Block Open-Sources Berd, a Cross-Model AI Agent Workspace

    Block has open-sourced Berd, a cross-model AI agent workspace, available on GitHub under the Apache 2.0 license.

    Why it mattersAn open-source agent workspace lowers cost to experiment, enabling smaller teams to pilot automation without vendor lock-in.

  9. ANALYSIS

    Backlash and ARR doubts, yet AI hiring stays broad-based

    Commentary warned of shaky economics and rising safeguards, but our tracker shows 4,124 open AI roles across 174 employers, with no surge in safety hiring despite watermarking momentum.

    Why it mattersDespite backlash narratives, hiring breadth persists, suggesting firms still prioritize builders over safety specialists.

  10. ANALYSIS

    Lots of safety talk, just 40 safety jobs in our tracker

    Commentators flagged pauses, surveillance and hype; employer job feeds show continued expansion in engineering and data, with minimal growth in evaluation and safety roles.

    Why it mattersSafety chatter outpaces openings, signaling limited near-term headcount for evaluation and risk roles.

  11. ANALYSIS

    Despite 'unraveling' claims, OpenAI keeps hiring at pace

    Commentators flag crisis, bubbles and productivity shocks; our employer listings show OpenAI and Anthropic still adding roles, safety hiring remains small, and broad demand is led by engineering and data.

    Why it mattersContinued expansion at leading labs anchors demand for engineers and data talent, with safety headcount still modest.

  12. ANALYSIS

    Safety alarms are loud; safety hiring is not

    Commentators flagged rising alignment risk and OpenAI turmoil, but our tracker shows builder-heavy hiring and OpenAI still opening roles faster than Anthropic.

    Why it mattersBuilder-heavy requisitions dominate, even as safety concerns rise, and OpenAI’s pace outstrips Anthropic’s.

  13. ANALYSIS

    Safety alarms grow; only 39 safety roles in 3,624 AI openings

    This week’s posts flag rising alignment risk and rapid adoption, while our tracker shows hiring remains concentrated in engineering and data, with few evaluation and safety roles.

    Why it mattersTiny safety share amid thousands of postings implies risk work lags adoption.

  14. ANALYSIS

    Engineering hires dwarf safety as hype swirls around AI

    Practitioners tout capability leaps, spending ceilings and trust crises; our 3,660-role tracker shows engineering demand surging, safety thin, and vendor hiring steady.

    Why it mattersSustained engineering demand, thin safety hiring, and steady vendors shape where skills pay off now.

All issues