{
  "schemaVersion": 2,
  "slug": "2026-07-16",
  "title": "AI Round-up: July 16, 2026",
  "date": "2026-07-16",
  "publishedAt": "2026-07-16T05:14:12Z",
  "sinceUtc": "2026-07-15T05:12:57Z",
  "rangeLabel": "Last 7 days",
  "summary": "OpenAI supplied the clearest safety signal in this snapshot. Its GPT-Red thread framed automated red teaming as a way to harden GPT-5.6 against prompt-injection attacks, while its policy post tied AI safety to state and federal action. Anthropic pushed the same control theme from a different angle with new agentic-misalignment simulations, and Microsoft added a concrete supply-chain security case through its AsyncAPI npm compromise analysis.\n\nThe rest of the window centred on agent tooling and infrastructure. OpenAI and OpenAIDevs promoted Codex Micro, Perplexity described SPACE as the sandbox platform behind Computer, GitHub brought Copilot progress into Jira, Databricks surfaced Thinking Machines Lab's Inkling model and data-native agents, and DeepMind argued that scientific agents are creating a new validation bottleneck.",
  "leadSignal": "GPT-Red safety flywheel",
  "topSignals": [
    {
      "label": "GPT-Red safety flywheel",
      "href": "https://x.com/OpenAI/status/2077446718728425686",
      "tag": "X"
    },
    {
      "label": "Agentic misalignment tests",
      "href": "https://x.com/AnthropicAI/status/2077452646303006927",
      "tag": "X"
    },
    {
      "label": "Agent runtime tooling",
      "href": "https://x.com/perplexity_ai/status/2077432518081744979",
      "tag": "X"
    }
  ],
  "topics": [
    "Anthropic",
    "OpenAI",
    "Codex",
    "Safety and security",
    "AI coding tools"
  ],
  "totalSignals": 40,
  "stats": {
    "news": 10,
    "x": 10,
    "youtube": 10,
    "reddit": 10
  },
  "paths": {
    "html": "/roundup/2026-07-16/",
    "json": "/roundup/2026-07-16.json",
    "markdown": "/roundup/2026-07-16.md"
  },
  "urls": {
    "html": "https://www.ai-roundup.dev/roundup/2026-07-16/",
    "json": "https://www.ai-roundup.dev/roundup/2026-07-16.json",
    "markdown": "https://www.ai-roundup.dev/roundup/2026-07-16.md"
  },
  "sections": [
    {
      "id": "news",
      "title": "News / Articles",
      "shortLabel": "News",
      "mobileTabLabel": "News",
      "count": 10,
      "items": [
        {
          "title": "The US is advancing AI safety through state and federal action",
          "source": "OpenAI",
          "publishedAt": "2026-07-15T16:37:34Z",
          "url": "https://openai.com/index/advancing-ai-safety-through-state-and-federal-action/"
        },
        {
          "title": "Unpacking the AsyncAPI npm supply chain compromise and import-time payload delivery",
          "source": "Microsoft",
          "publishedAt": "2026-07-16T01:36:21Z",
          "url": "https://www.microsoft.com/en-us/security/blog/2026/07/15/unpacking-asyncapi-npm-supply-chain-compromise-import-time-payload-delivery/"
        },
        {
          "title": "JPMorgan CEO Dimon says Anthropic's Mythos AI risks are a 'real issue'",
          "source": "Reuters",
          "publishedAt": "2026-07-16T00:24:00Z",
          "url": "https://www.reuters.com/business/finance/jpmorgan-ceo-dimon-says-anthropics-mythos-ai-risks-are-real-issue-2026-07-16/"
        },
        {
          "title": "Inkling model from Thinking Machines Lab now on Databricks",
          "source": "Databricks",
          "publishedAt": "2026-07-15T19:15:22Z",
          "url": "https://www.databricks.com/blog/inkling-thinking-machines-lab-now-databricks"
        },
        {
          "title": "AI startup Thinking Machines launches an open-weight AI model",
          "source": "Reuters",
          "publishedAt": "2026-07-15T18:03:44Z",
          "url": "https://www.reuters.com/business/ai-startup-thinking-machines-launches-an-open-weight-ai-model-2026-07-15/"
        },
        {
          "title": "How AI Agents are transforming scientific discovery",
          "source": "Google DeepMind",
          "publishedAt": "2026-07-15T15:26:04Z",
          "url": "https://deepmind.google/public-policy/conjecture-machines-ai-agents-and-the-new-validation-bottleneck-in-science/?utm_source=linkedin&utm_medium=social&utm_campaign=&utm_content="
        },
        {
          "title": "Data-Native AI Agents: Why Agents Must Move to Your Data",
          "source": "Databricks",
          "publishedAt": "2026-07-15T18:00:44Z",
          "url": "https://www.databricks.com/blog/data-native-ai-agents-why-agents-must-move-your-data"
        },
        {
          "title": "From Answers to Action: What’s New with Copilot in Dynamics 365 Customer Service",
          "source": "Microsoft",
          "publishedAt": "2026-07-15T15:00:00Z",
          "url": "https://www.microsoft.com/en-us/dynamics-365/blog/it-professional/2026/07/15/service-agent-microsoft-365-copilot-customer-service/"
        },
        {
          "title": "Airbus picks Iliad's Scaleway for AI, defence work in sovereignty push",
          "source": "Reuters",
          "publishedAt": "2026-07-16T05:02:00Z",
          "url": "https://www.reuters.com/legal/litigation/airbus-picks-iliads-scaleway-ai-defence-work-sovereignty-push-2026-07-16/"
        },
        {
          "title": "Nvidia partners with Japan robotics firms on AI development",
          "source": "Reuters",
          "publishedAt": "2026-07-16T04:56:00Z",
          "url": "https://www.reuters.com/business/media-telecom/nvidia-partners-with-japan-robotics-firms-ai-development-2026-07-16/"
        }
      ]
    },
    {
      "id": "youtube",
      "title": "YouTube",
      "shortLabel": "Video",
      "mobileTabLabel": "YouTube",
      "count": 10,
      "items": [
        {
          "title": "Introducing the Codex Micro",
          "channel": "OpenAI",
          "publishedAt": "2026-07-15T19:42:38Z",
          "views": 64211,
          "url": "https://www.youtube.com/watch?v=m8uUUUsMD3Y"
        },
        {
          "title": "OpenAI vs Anthropic",
          "channel": "Matthew Berman",
          "publishedAt": "2026-07-15T17:31:52Z",
          "views": 48975,
          "url": "https://www.youtube.com/watch?v=nxr20lcqagg"
        },
        {
          "title": "Codex and Claude Shipped Browser Updates. This Changes Everything.",
          "channel": "Riley Brown",
          "publishedAt": "2026-07-15T13:00:36Z",
          "views": 18797,
          "url": "https://www.youtube.com/watch?v=juPDqb89dew"
        },
        {
          "title": "Claude Just Changed Completely: Here's How It Works (In 2026)",
          "channel": "Paul J Lipsky",
          "publishedAt": "2026-07-15T13:00:00Z",
          "views": 17713,
          "url": "https://www.youtube.com/watch?v=bj04doEDOY4"
        },
        {
          "title": "Anthropic Now Says AI Could Kill Us All...",
          "channel": "AI Revolution",
          "publishedAt": "2026-07-15T22:58:42Z",
          "views": 8376,
          "url": "https://www.youtube.com/watch?v=8D0INXhxUIw"
        },
        {
          "title": "Complete Guide to ChatGPT 5.6 + Prompting Guide",
          "channel": "Matt Wolfe",
          "publishedAt": "2026-07-15T22:17:52Z",
          "views": 9149,
          "url": "https://www.youtube.com/watch?v=MDy_b9F7oUc"
        },
        {
          "title": "How to Use Claude Code as an Agency and Actually Make Money",
          "channel": "Mikey No Code",
          "publishedAt": "2026-07-15T14:15:16Z",
          "views": 9375,
          "url": "https://www.youtube.com/watch?v=mUkDBxwMZ_c"
        },
        {
          "title": "Fable + Sol is a CHEAT CODE for Claude Code (Cheaper, more Powerful)",
          "channel": "Jay E | RoboNuggets",
          "publishedAt": "2026-07-15T07:04:15Z",
          "views": 8517,
          "url": "https://www.youtube.com/watch?v=4kRdt18_dFY"
        },
        {
          "title": "OpenAI Codex Hackathon is Live | Register for FREE",
          "channel": "Akshay Saini",
          "publishedAt": "2026-07-15T05:30:29Z",
          "views": 6661,
          "url": "https://www.youtube.com/watch?v=YVGTkvX3RL8"
        },
        {
          "title": "How Anthropic Engineers ACTUALLY Automate Their Work",
          "channel": "Austin Marchese",
          "publishedAt": "2026-07-15T15:45:14Z",
          "views": 3076,
          "url": "https://www.youtube.com/watch?v=3Y6kVI3fR9c"
        }
      ]
    },
    {
      "id": "reddit",
      "title": "Reddit",
      "shortLabel": "Reddit",
      "mobileTabLabel": "Reddit",
      "count": 10,
      "items": [
        {
          "title": "Why doesn't Cloud Code have a fork feature(the /branch function sucks)",
          "subreddit": "ClaudeCode",
          "publishedAt": "2026-07-15T16:35:05Z",
          "score": 0,
          "comments": 17,
          "url": "https://www.reddit.com/r/ClaudeCode/comments/1uxb2cp/why_doesnt_cloud_code_have_a_fork_featurethe/"
        },
        {
          "title": "Best VS code /github copilot agent?",
          "subreddit": "GithubCopilot",
          "publishedAt": "2026-07-15T21:43:48Z",
          "score": 0,
          "comments": 11,
          "url": "https://www.reddit.com/r/GithubCopilot/comments/1uxjdew/best_vs_code_github_copilot_agent/"
        },
        {
          "title": "When you have both a codex Pro and claude code Max account... no sleep I guess",
          "subreddit": "ClaudeCode",
          "publishedAt": "2026-07-16T04:38:24Z",
          "score": 0,
          "comments": 6,
          "url": "https://www.reddit.com/r/ClaudeCode/comments/1uxspvd/when_you_have_both_a_codex_pro_and_claude_code/"
        },
        {
          "title": "I built an open-source way to control GitHub Copilot agents from my phone",
          "subreddit": "GithubCopilot",
          "publishedAt": "2026-07-16T00:01:45Z",
          "score": 0,
          "comments": 5,
          "url": "https://www.reddit.com/r/GithubCopilot/comments/1uxmu94/i_built_an_opensource_way_to_control_github/"
        },
        {
          "title": "I build VibeKeys — after the Codex Micro reaction in this sub, here's the $99 keypad I've been building with a Codex mode",
          "subreddit": "codex",
          "publishedAt": "2026-07-16T05:02:02Z",
          "score": 0,
          "comments": 2,
          "url": "https://www.reddit.com/r/codex/comments/1uxt69w/i_build_vibekeys_after_the_codex_micro_reaction/"
        },
        {
          "title": "Both Claude and OpenAI Reset Their Limits",
          "subreddit": "Anthropic",
          "publishedAt": "2026-07-16T05:07:33Z",
          "score": 0,
          "comments": 2,
          "url": "https://www.reddit.com/r/Anthropic/comments/1uxta09/both_claude_and_openai_reset_their_limits/"
        },
        {
          "title": "How can I incorporate the Copilot AI agent into an MS Power Automate process?",
          "subreddit": "PowerAutomate",
          "publishedAt": "2026-07-15T13:20:43Z",
          "score": 0,
          "comments": 2,
          "url": "https://www.reddit.com/r/PowerAutomate/comments/1ux5otb/how_can_i_incorporate_the_copilot_ai_agent_into/"
        },
        {
          "title": "I built a Telegram bridge for OpenAI Codex CLI: live models, MCP, skills, subagents, sessions, diffs, and safe account rotation",
          "subreddit": "codex",
          "publishedAt": "2026-07-15T11:44:45Z",
          "score": 0,
          "comments": 0,
          "url": "https://www.reddit.com/r/codex/comments/1ux3hoi/i_built_a_telegram_bridge_for_openai_codex_cli/"
        },
        {
          "title": "Any reason to use claude co-work rather than just move to claude code?",
          "subreddit": "ClaudeAI",
          "publishedAt": "2026-07-16T04:44:19Z",
          "score": 0,
          "comments": 4,
          "url": "https://www.reddit.com/r/ClaudeAI/comments/1uxstye/any_reason_to_use_claude_cowork_rather_than_just/"
        },
        {
          "title": "Claude code mistakes to avoid",
          "subreddit": "ClaudeCode",
          "publishedAt": "2026-07-16T04:46:58Z",
          "score": 0,
          "comments": 1,
          "url": "https://www.reddit.com/r/ClaudeCode/comments/1uxsvs3/claude_code_mistakes_to_avoid/"
        }
      ]
    },
    {
      "id": "x",
      "title": "X / Twitter",
      "shortLabel": "X",
      "mobileTabLabel": "X",
      "count": 10,
      "items": [
        {
          "text": "Introducing GPT-Red\nAn internal automated red teamer on a mission to find our models’ prompt injection vulnerabilities at scale, helping us build stronger defenses before wider deployment.\nGPT-Red: Unlocking Self-Improvement for Robustness\nFrom openai.com",
          "handle": "OpenAI",
          "publishedAt": "2026-07-15T17:34:29Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/OpenAI/status/2077446718728425686"
        },
        {
          "text": "Training against GPT‑Red makes GPT‑5.6 substantially more resilient. To measure this, we replayed some of GPT‑Red’s strongest attacks—none of which our models had seen during training. GPT‑5.6 Sol proved to be our most robust model against prompt injections to date, with 6× fewer",
          "handle": "OpenAI",
          "publishedAt": "2026-07-15T17:34:30Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/OpenAI/status/2077446722683650525"
        },
        {
          "text": "AI agents are already being used to improve the capabilities of our next-generation models.\nWe believe with GPT-Red that we have started to unlock a similar flywheel for safety, where today's models can be used to make tomorrow's models more robust, aligned, and trustworthy.",
          "handle": "OpenAI",
          "publishedAt": "2026-07-15T17:34:30Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/OpenAI/status/2077446723992228167"
        },
        {
          "text": "New Anthropic research: Agentic misalignment in Summer 2026.\nA year after our blackmail experiments, we found four more ways that today’s autonomous AI agents misbehave in simulations.\nRead more: alignment.anthropic.com/2026/agentic-m…",
          "handle": "AnthropicAI",
          "publishedAt": "2026-07-15T17:58:02Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/AnthropicAI/status/2077452646303006927"
        },
        {
          "text": "We tested many AI models, including Claude, in the four scenarios. Even though these weren’t real incidents, they demonstrate clear misaligned behavior that should be studied further and mitigated.\nFind all the transcripts from the scenarios here: aenguslynch.com/portfolio-tran…",
          "handle": "AnthropicAI",
          "publishedAt": "2026-07-15T17:58:03Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/AnthropicAI/status/2077452649000042614"
        },
        {
          "text": "Meet kbd-1.0-codex-micro, built with @work_louder.\nMap the buttons and joystick to your workflow, and keep your pinned chats in view.\nGet yours before stock returns 410.\n02:13",
          "handle": "OpenAIDevs",
          "publishedAt": "2026-07-15T16:12:07Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/OpenAIDevs/status/2077425991790870644"
        },
        {
          "text": "Introducing SPACE, the sandbox platform behind Perplexity Computer.\nIt creates isolated environments for code, files, and long-running agent sessions.\nSPACE has handled 100% of Computer production traffic since June.\nMaking SPACE: Secure and Efficient Runtimes for Long-Running Agents\nFrom research.perplexity.ai",
          "handle": "perplexity_ai",
          "publishedAt": "2026-07-15T16:38:03Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/perplexity_ai/status/2077432518081744979"
        },
        {
          "text": "GitHub Copilot for @Atlassian's Jira is now generally available.\nWith Copilot in Jira, you can:\n• stream agent progress to Jira\n• provide follow-up instructions directly in the Jira chat panel\n• connect your GitHub org and repos to the Jira app faster with fewer steps at the\n01:40",
          "handle": "github",
          "publishedAt": "2026-07-15T21:00:37Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/github/status/2077498593426985056"
        },
        {
          "text": ".@thinkymachines' first open-weights model, Inkling, is now available on Databricks through Unity AI Gateway.\nAs a day zero launch partner, Databricks gives enterprise teams access to a model that excels at coding and agentic reasoning and supports multimodal inputs.\nTeams can",
          "handle": "databricks",
          "publishedAt": "2026-07-15T18:31:06Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/databricks/status/2077460968125145509"
        },
        {
          "text": "We've open-sourced Grok Build and have reset usage limits for all users.\nOpen sourcing Grok Build allows anyone to support making a reliable and robust harness. Check out our code, including the Git repo for the Grok Build CLI.\nSpaceXAI — Creators of Grok, the AI Chatbot\nFrom x.ai",
          "handle": "SpaceXAI",
          "publishedAt": "2026-07-15T20:44:29Z",
          "views": null,
          "likes": 0,
          "url": "https://x.com/SpaceXAI/status/2077494535387828644"
        }
      ]
    }
  ]
}