Skip to main content
AmigoFish

YPO Technology Network AI Brief

Stephen ForteAudio

Format:
Podcast
New episodes:
Daily
Typical length:
about 7 minutes
Latest episode:
Not rated yet
CosmosOpen at the sourceReport a problem

About this show

AI moves fast. Your briefing should move faster. The YPO Technology Network AI Brief is a daily breakdown of the AI developments that actually matter to your business. No hype, no jargon, no filler — just what changed, what it costs you or saves you, and what to tell your team on Monday. Hosted by Stephen Forte for the leaders who don't have time to chase the news but can't afford to miss it.

Show details and episodes come from the publisher's public feed at media.rss.com. Last updated . Names, artwork and text belong to their publishers.

Cosmos

Similar by topic and creator

See the full Cosmos

Episodes

  • AI Vendor Risk: When Outlook Goes Down, Your Agents Go With It

    · 6 min

    Microsoft's Outlook and much of Microsoft 365 went down late Thursday 12 November and were not fully restored, by Microsoft's public account, until Saturday morning 14 November. Microsoft says this was a technical issue, not a cyberattack and not AI, from a problem in an earlier network capacity upgrade that showed up after a recent network change. The CEO lesson is that agents sitting on Outlook, the calendar, and Microsoft Graph went down with that pipe. Key takeaways. Have a way to reach people that does not run through Outlook or Teams, and test it. List every mail and calendar agent and what it should do when Graph times out. A service credit is bookkeeping, not a recovery plan. Decide ahead of time how long you will wait before you fail over, knowing a DNS change can take an hour. Checklist (as spoken). You do not need to write this down while you listen. - A way to reach staff and key customers that does not run through Outlook or Teams. And test it. A phone tree, a text, or a second messaging tool is enough. The test is the part most companies skip. For us, Slack covered the inside of the company. It did not cover customers. Why: most customer communication still goes through email. - A short list of every agent or workflow that reads mail, sends mail, or touches the calendar through Graph. Write down what each one should do when those requests time out. Stop and alert a person. Queue the work. Or fail closed. An agent that keeps retrying in the dark is worse than an agent that stops. Why: Graph requests may time out, and an agent can keep running with nothing useful to work with. - Know what the contract actually pays you when the service is down. If you are on a paid business plan, you can ask for a service credit. It is a reduction on a future bill for the service that was down. It is not automatic, and it is not payment for lost work. You have to show that your own users were hit, and you have to ask by the end of next month. Treat that as bookkeeping, not a recovery plan. If email is how you sell, book, or approve work, you need a second path, not a better invoice credit. Why: the credit will not send the stuck messages. The plan is the second path, and how long you will wait before you use it. Timeline (Dublin / US Eastern). 12 Nov, 17:54 Dublin / 12:54 p.m. ET: EX1550543 opened. 18:35 / 1:35 p.m. ET: MO1550821. 13 Nov, 16:40 Dublin / 11:40 a.m. ET: preliminary cause. 14 Nov, 06:09 Dublin / 1:09 a.m. ET: customers restored, 36 hours later. Admin-center recaps list a different Friday end time than the Saturday public post used on air. Terms. Microsoft Graph: the interface agents use for mail, calendar, and contacts. DNS failover: changing where your mail is delivered, which can take an hour. SLA credit: a future-bill reduction on the down service if you ask by the end of next month. Examples named. Our company kept a Gmail standby ready in about fifteen minutes and did not flip, because nothing critical needed it and DNS can take an hour. Bots still ran but not off incoming mail. Slack was fine. OpenAI and Anthropic Microsoft 365 connectors failed. ChatGPT Wallet launched 13 Nov. OpenAI Control Room stops research-system work within 35 seconds and fails closed. Sources: First acknowledgment (https://x.com/MSFT365Status/status/2120938282204688468); Cause statement (https://x.com/Microsoft/status/2121282013600907325); Saturday restoration (https://x.com/MSFT365Status/status/2121485746334777593); Wallet (https://openai.com/index/introducing-chatgpt-wallet/); Control Room (https://openai.com/index/control-room/). The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • AI Computer Use: Connectors First, Screens Last

    · 13 min

    Can your AI actually use your computer? Claude, ChatGPT, and Grok Bot all can, but there are four very different ways an agent touches your systems. Stephen Forte explains why legacy work still needs computer and browser control, compares the methods, and gives one rule. Connectors first, page or app structure next, screens last. In this episode: - Why this still matters. A lot of real work still lives in systems with no connector: carrier partner portals, point of sale back offices, state payroll tax sites, and old Windows insurance applications. - Four methods, most precise to least. Connectors and MCP servers first. Browser structure next, with tools such as Playwright and agent-browser. Then the operating system accessibility tree, the method behind agent-desktop. Screenshots and mouse coordinates last, the slowest and most expensive option. - agent-desktop and agent-browser. Open source tools from Vercel Labs that read app or page structure instead of pictures of the screen. agent-desktop is a set of hands and eyes any agent can call. It is getting better, and it can still be slow. - Community add-ons. Windows-MCP, Peekaboo, and computer-use-linux bolt the same accessibility features onto Windows, Mac, and Linux. Worth a look, not a recommendation. - What each assistant can do today. Claude has computer use, browser use, Claude Code, and Claude in Chrome. After Claude Fable 5.2, the older computer use API tool now returns errors everywhere except Amazon Bedrock. Desktop computer control is still a research preview, not on Team or Enterprise plans yet. ChatGPT product names have changed fast: Operator, agent mode, Atlas, then Work. What exists now is ChatGPT Work and the desktop app's Computer Use. Grok Bot works from its own cloud computer, with connected apps preferred over clicking. - The rule, plus the risk. Pick the most precise method that reaches the work. Do not run a screen agent on a laptop that has client email open. A signed-in browser session is a long-lived credential. Retest after vendor updates and site redesigns. Sources: - min heo on X, the agent-desktop tip: X post, Nov 9, 2026 (https://x.com/minheo93698092/status/2119765548204515549) - Vercel Labs: agent-desktop (https://github.com/vercel-labs/agent-desktop) and agent-browser (https://agent-browser.dev) - Anthropic: Computer use tool (https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool) - Anthropic: What's new in Claude Fable 5.2 (https://platform.claude.com/docs/en/models/fable-5-2/whats-new-fable-5-2) - Anthropic: Claude Code computer use (https://code.claude.com/docs/en/computer-use) and Claude in Chrome generally available (https://claude.com/resources/articles/claude-in-chrome-generally-available) - OpenAI: ChatGPT Atlas retirement (https://help.openai.com/en/articles/20001371-chatgpt-atlas-retirement), desktop browser (https://help.openai.com/en/articles/20001277-using-the-built-in-browser-in-the-chatgpt-desktop-app), and API computer use (https://developers.openai.com/api/docs/guides/tools-computer-use) - xAI: Grok Bot computer and apps (https://docs.x.ai/grok-bot/computer-and-apps) - Community add-ons: Windows-MCP (https://github.com/CursorTouch/Windows-MCP), Peekaboo (https://github.com/openclaw/Peekaboo), and computer-use-linux (https://github.com/agent-sh/computer-use-linux) The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • AI Agent Testing: The Business Owner Defines Good

    · 7 min

    Public AI benchmarks cannot tell you which agent to trust on your workflows. The business owner should define good, test the outcome first, then every step, and when a build looks finished, force a visual workflow and a full process map. Key takeaways. The business owner writes the tests; engineering should not invent success alone. Test the outcome first, then each step. When it appears to work, force the guide, visual workflow, and process map. Check new models, including Gemini 4 Carbon, on your own cases. Rerun tests after every update. Fund checking before more output. Monday checklist (as spoken). You do not need to write this down while you listen. - Outcome tests owned by the business. The owner writes what good and bad finished results look like. Why: they know what good looks like, not only what is easy to automate. - Step tests across variations. Change the inputs. Require intermediate values and sources. Score logic breaks between steps. Why: a good last message can hide a broken middle step. - Decline and escalate cases. Include the cases where the right answer is no, not enough information, or send it to a person. Score an unjustified action separately from unfinished work. Finished is not the same as right. That is the light version of this week's Visa lesson, without importing a consumer shopping score into your shop floor. Why: completion is not correctness. - The same case, repeated. Run an identical case several times. Consistency is a separate score from a single lucky pass. Why: one pass is not consistency. - Compare against past human-handled cases. Take closed work a competent person already decided. Blind the agent. Compare outcomes and the follow-ups required. Why: your closed work is local ground truth. - Finished-product map, then rerun after every vendor or model update. Require the guide, the visual workflow, and the full process map after it appears to work. Then treat every platform change as a release candidate and rerun the suite. We learned that the hard way when a tool update broke multi-step workflows my team relied on. Memory kept outside the tool let us move. The test suite is how you notice the change in minutes, not after customers do. Why: leftover paths and vendor updates show up here. Examples named. Invoice, claim, and customer-change cases (types, not industry claims). Leftover data writes or folder references. Gemini 4 Carbon's benchmark claims. Elvis Saravia and Lauren (9 Nov). Meta AutoRefine. DoorDash's Ask DoorDash harness (hidden tool path only). Terms. Public benchmark: a published model test, not your workflow. Intermediate values: mid-process numbers, materials, and sources. Blind the agent: hide the prior human decision. Grader: your scoring method; AutoRefine found older graders go stale. Sources: Elvis Saravia, 9 Nov (https://x.com/omarsar0/status/2119795295630324042); Lauren, 9 Nov (https://x.com/poteto/status/2119824409087348984); Google, Gemini 4 Carbon (https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-carbon-now-available); Meta, AutoRefine (https://arxiv.org/abs/2611.09291); DoorDash evaluation (https://careersatdoordash.com/blog/building-ask-doordash-part-three-evaluation/); Visa benchmark release (https://investor.visa.com/news/news-details/2026/Visa-Releases-First-of-Its-Kind-Industry-Benchmark-for-AI-Shopping-Almost-Half-of-Test-Runs-Fail/); Visa whitepaper (https://www.visapaymentsfrontier.io/articles/agentic-commerce-benchmark/assets/visa-agentic-commerce-benchmark-whitepaper.pdf). The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • AI ROI Metrics: Stop Counting Hours Saved, Start Counting Growth

    · 7 min

    New Wharton and Deloitte findings show most companies still prove AI value with hours saved, while teams that also track growth report stronger returns. Stephen Forte explains why hours saved is capacity, not cash, and what should go on a one-page AI scorecard. Plus how long it can take to actually shut an agent off. In this episode: - The wrong headline. Hours saved is the most common AI metric among managers who measure return. It is capacity, not cash. If that capacity is not allocated to something, it gets reabsorbed. - Wharton and GBK, October 2026. Departments that track at least one growth measure, such as net-new revenue, conversion, output, or customer satisfaction, report stronger returns than those tracking efficiency alone. These are survey answers, not audited financials. Organizations on one approved platform, and those with a designated chief AI officer, also report stronger returns. - Deloitte, October 29. Most leaders report faster decisions and execution. Only a quarter say they use AI to deeply transform the business. Most companies now track the total AI bill. Far fewer track token volume or the cost of a unit of work. Nearly one in four say accountability is unclear when AI gets it wrong. - West Monroe, via CIO Dive on November 9. A large majority of U.S. tech executives now put more than ten percent of IT budget into AI, a sharp rise from last year. - A one-page scorecard. Make a growth or outcome measure the headline. Treat token and AI spend like any other business expense, with a purpose and a deliverable. Reinvent the work before you cut. Put a named owner on the measurement and standardize on one approved platform. Leaders should report the same way they ask the organization to report. - Lightning item. An AWS test with Okta found that deleting the agent's resource connection refused new requests immediately, while deactivating the agent sat in a beta API queue for eighty-six minutes. Tokens already issued kept working up to five minutes. This was vendor-published, one test organization, and sample code marked non-production. Ask your identity team which switch actually stops an agent, and test that plan once before you need it. Sources: - Wharton Human-AI Research and GBK Collective, 2026 AI Adoption Report (Oct 2026) (https://knowledge.wharton.upenn.edu/special-report/2026-ai-adoption-report/) and PDF (https://knowledge.wharton.upenn.edu/wp-content/uploads/2026/10/Wharton-GBK-AI-Adoption-Report-2026.pdf) - GBK press summary (https://www.gbkcollective.com/thoughtleadership/wharton-gbk-ai-study-2026-press-release) - Deloitte, State of AI in the Enterprise (Oct 29, 2026) (https://www.deloitte.com/us/en/about/press-room/enterprise-ai-enters-an-roi-reckoning.html) - West Monroe via CIO Dive / Yahoo Finance (Nov 9, 2026): CIOs upping budgets to meet AI (https://finance.yahoo.com/technology/ai/articles/cios-upping-budgets-meet-ai-120000080.html) - AWS Machine Learning Blog: Okta agent access, part 1 (https://aws.amazon.com/blogs/machine-learning/govern-ai-agent-access-with-okta-cross-app-access-and-amazon-bedrock-agentcore-part-1/) and timing results, part 2 (https://aws.amazon.com/blogs/machine-learning/govern-ai-agent-access-with-okta-cross-app-access-and-amazon-bedrock-agentcore-part-2/) The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • AI Agent Controllability: Check Before The Agent Acts

    · 6 min

    Most companies still find out what their AI agents did by reading a log after the fact. Stephen Forte explains controllability: a check that runs before the agent acts and can say no, five working guidelines, and a trend in which Mistral and ANZ are building the same kind of controls. In this episode: - Controllability defined. A check that runs before the agent acts and decides whether the action is allowed. If the check says no, the action does not happen. A log afterward is not controllability. - Anthropic inference hooks. Now generally available on Claude Enterprise. Before Claude answers a prompt or runs a tool call, your company's security or data loss prevention server gets an allow or deny vote, inside Claude's infrastructure. One setup covers chat, Claude Code, Cowork, and connected tools. This is an architecture lesson, not a buying recommendation. - Five working guidelines. A check before the agent acts. Short-lived credentials, with tokens that expire in minutes. Memory, ledgers, rules, and procedures kept outside the tool in files you own. Subject experts own the workflow, and a human stays accountable for high-stakes outcomes. Retest after every vendor or model update, and keep a dependency inventory. - Four CISO questions. What untrusted content the agent reads, what it can do and on whose identity, the blast radius if it goes wrong, and whether you can see its actions in your logs. Least agency means the narrowest capability that still finishes the job. - Mistral Studio. Workflows with human approvals are generally available. Managed agents are in public preview. OpenTelemetry traces show model calls, tool use, and approvals. Private cloud and on-premises timing for the new pieces is reported as expected by late November, not as something already shipped. - ANZ Control Plane. CEO Nuno Matos told investors the bank is building an Enterprise AI Control Plane for visibility and technical control, with defined limits and a way to stop activity outside those limits. Humans must remain in control. - Lightning item. Anthropic's incident response agent stayed inside human review when a newer model started reaching out to another agent mid-incident, because the write surface was bounded. Design the check for where the model will be in six months. Sources: - Anthropic: Claude Enterprise security updates, inference hooks generally available (https://claude.com/resources/articles/claude-enterprise-security-updates) - Anthropic Help Center: Inference hooks overview (https://support.claude.com/en/articles/16059458-inference-hooks-overview) - Anthropic platform docs: Inference hooks (https://platform.claude.com/docs/en/manage-claude/inference-hooks) - Anthropic: CISO guide to agentic AI (https://claude.com/resources/articles/ciso-guide-to-agentic-ai) - Anthropic: Applying Zero Trust to agentic AI services (https://claude.com/resources/guides/zero-trust-for-ai-agents/applying-zero-trust-to-agentic-ai-services) - Mistral: Mistral Studio announcement (https://mistral.ai/news/build-with-studio/) - Mistral: Release notes, Managed Agents public preview (https://docs.mistral.ai/resources/release-notes?entry=2026-11-05-a528ab2e5db6) - ANZ: CEO Nuno Matos FY26 investor remarks (https://www.anz.com.au/newsroom/media/2026/november/anz-ceo-nuno-matos-remarks-investor-briefing-anz-2026-full-year-results/) The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • AI Agent Reliability: Test Wrong Decisions, Not Just Completed Tasks

    · 7 min

    Visa's new AI shopping benchmark found agents that failed usually still chose to buy. Stephen Forte explains why CEOs should test agents on the cases where the right answer is to decline or ask, and why memory kept outside the tool matters when a vendor update breaks. In this episode: - Visa's Agentic Commerce Benchmark. Twenty-two models from twelve labs ran fifteen shopping scenarios three times each, for nine hundred and ninety runs. Overall pass rate fifty-seven point two percent; only thirty-four point two percent of model and scenario pairs passed all three attempts. - Caveats that matter. Simulated market and customer, stopped before checkout, no money moved, a small scenario set, and Visa is an interested party. Read it for how it grades, not as a score for your business. - How a run passes. An agent had to discover every customer requirement and then make the right call, including choosing nothing when nothing fits. Guessing the right product without asking the right questions could still fail. - What failed agents did. In fifty-four percent of the failed runs, the model still chose to buy. That is half of the failures, not half of all runs. Visa also counted false refusals as failures. - Completion is not correctness. Most agent pilots are graded on whether the task finished. Demos show the cases that go well. The cases that matter are where the right answer is no. - Three tests before you trust an agent. Mix in decline-or-ask cases, rerun identical cases for repeatability, and score acting-when-it-shouldn't separately from refusing-when-it-should. - Grok Bot update. A first-hand account of multi-step stalls, faster token burn, worse judgment and a false self-upgrade claim after a vendor update. Public context from Elon Musk and Matt Palmer on November eighth, without a named cause or timeline. - Resilience outside the tool. Client work moved to Claude in about fifteen minutes because memory lived in backed-up ledgers. Keep a dependency inventory for tools with their own Linux machine. Retest agents after every vendor update. Sources: - Visa, Nov 5: Agentic Commerce Benchmark release (https://investor.visa.com/news/news-details/2026/Visa-Releases-First-of-Its-Kind-Industry-Benchmark-for-AI-Shopping-Almost-Half-of-Test-Runs-Fail/) - Visa whitepaper: Methodology and grading (https://www.visapaymentsfrontier.io/articles/agentic-commerce-benchmark/assets/visa-agentic-commerce-benchmark-whitepaper.pdf) - Elon Musk on X, Nov 8: Grok Bot fixes underway (https://x.com/elonmusk/status/2119419126045454531) - Matt Palmer on X, Nov 8: User feedback request (https://x.com/mattyp/status/2119474599708827957) - xAI: Grok Bot product description (https://x.ai/news/introducing-grok-bot) The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • AI and Organizational Restructuring: Redesign the Work Before Cutting Jobs

    · 7 min

    AI exposes the layers, hand-offs and busy work inside a company. Stephen Forte explains why leaders should redesign the work before cutting jobs or flattening the organization. In this episode: - Three restructuring announcements. NiCE, Canva and SUSE named AI among their reasons for change, alongside other business pressures. These are not simply AI layoffs. - Readiness exposes waste. Preparing for AI reveals weak data, inappropriate access and the hand-offs between a request and a result. - Double clutter. Email and Excel create work on both sides of a request. Count the relays before changing the org chart. - Sequence the change. A Harvard Business School AI Institute working paper links AI capabilities with flatter hierarchies and wider managerial spans. The effects were stronger where people had the skills to use the technology; they were not automatic. - Evolution, not revolution. Reinvent the work first, then decide which layers to remove and how to organize the people you need. - Operational expertise matters. The people who understand accounting, sales or marketing can tell an agent what good looks like. Tool knowledge alone is not enough. - Train people and remove friction. Investment in training signals confidence in the team. Start by automating the spreadsheets and recurring reports people hate, giving them back the work they were hired to do. - One practical starting point. Pick one process, count its hand-offs and emails, automate the frustrating part, then reassess the layers. Sources: - NiCE, Nov 4: Quarterly results and restructuring announcement (https://www.nice.com/press-releases/nice-exceeds-revenue-and-eps-guidance-ranges-reporting-8-year-over-year-revenue-growth-in-third-quarter-2026) - Calcalist, Nov 4: NiCE workforce and financial context (https://www.calcalistech.com/ctechnews/article/6apcspzm8) - TechCrunch, Nov 4: Canva restructuring (https://techcrunch.com/2026/11/04/canva-lays-off-over-1200-people-amid-slowing-growth/) - The Register, Nov 2: SUSE restructuring (https://www.theregister.com/software/2026/11/02/suse-to-cut-one-in-ten-employees-because-of-ai/5306950) - Harvard Business School AI Institute, Nov 3: The Future is Flat (https://aiinstitute.hbs.edu/the-future-is-flat/) The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Who Gets The Power Tools?

    · 7 min

    Goldman Sachs just limited who can use its strongest coding agents. Stephen Forte asks who should get those power tools in your company. In this episode, Stephen Forte covers: - Goldman's October change. Bloomberg (Nov 2): Anthropic's Claude Code and Microsoft's GitHub coding tools are now for engineers. Traders and bankers lost access; the quant "strats" are exempt. Spokesperson Nick Carcaterra said that after experimenting on productivity and risk, they put the tools with those best suited to leverage them long term. - A real shift. In July, CIO Marco Argenti had praised vibe coding by traders and bankers who built tools and handed them to one of 12,000 engineers to polish. Earlier in 2026, Goldman had already restricted Claude for Hong Kong staff over confidential data going to third-party AI tools. - Respect the bank. Stephen will not question Goldman's call. Banking's regulatory environment is unique. Pulling coding agents back to engineers after experimentation is risk management where a bad tool can leave with client data. - Most companies are not Goldman. Builders outside engineering should get coding agents, 100%. Tools should go first to subject matter experts, who truly understand the problem. - IT's job. An engineer fixing accounting without understanding accounting is a mistake. IT should support subject matter experts, lift their skills and manage their code inside the company's accepted stack. - The real risk. Neutering your most experienced people and denying them modern tools. Those people will leave for a place that will give them the tools. - Stay inside the approved stack. Non-confidential prototyping elsewhere is usually fine. Operate inside tools the company already approved. Even Microsoft Copilot is getting better every day. - Replit's counterpoint. CEO Amjad Masad said agents on employee laptops are a systemic security risk and the ban hammer is coming down. Control what agents can touch and where they run; don't confuse that with locking out subject matter experts. - Juniors and judgment. Partner Chris Churchman warned heavy AI use can degrade reasoning. Stephen's view: juniors learn through structured use of AI. AI does busy work and calculations; people own judgment. A large language model cannot replace human judgment yet. - Tokens as expenses. Garry Tan of Y Combinator spends $100K to $200K a year on tokens. Stephen's rule: no hard caps by role. A gate for everybody through management, with deliverables that show value, same as justifying a $2,000 business trip. Sources: - Bloomberg Law (Nov 2, 2026): https://news.bloomberglaw.com/capital-markets/goldman-restricts-traders-and-bankers-from-using-some-ai-tools - Quartz (Nov 2, 2026): https://qz.com/goldman-sachs-restricts-ai-coding-tools-traders-bankers-110226 - CNBC, Chris Churchman (Aug 24, 2026): https://www.cnbc.com/2026/08/24/goldman-sachs-ai-partner-danger-skills.html - Tech in Asia, Goldman Hong Kong Claude restriction: https://www.techinasia.com/news/goldman-sachs-restricts-ai-coding-tools-engineers - The New Stack, Amjad Masad (Nov 3, 2026): https://thenewstack.io/ai-coding-agents-security/ - TechCrunch, Garry Tan tokens (Nov 2, 2026): https://techcrunch.com/2026/11/02/y-combinators-garry-tan-uses-ai-agents-so-much-he-spends-up-to-200k-a-year-on-tokens/ The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Write The Rules Down For The Machine

    · 6 min

    An agent can't follow a rule it can't read, and at most companies the rules live in PDFs, shared drives and people's heads. Stephen Forte walks through the step most companies skip before handing work to an agent: writing one process's rules down, in one place a machine can read. In this episode, Stephen Forte covers: - Machine-Readable Government. On Nov 2, Abu Dhabi and OpenAI rewrote a defined set of policies so an AI system can check a case against the rules and link each recommendation to the exact clause. An official verifies, corrects or overrides, and a person makes the final decision. - The first pilot. Food import registrations at Abu Dhabi's food safety authority: about 20,000 a year, drawing on 1,400 policy documents, some over 700 pages. Routine checks that took days now take minutes. Construction permits, licensing, HR compliance and procurement are next. - The caveat. It is a self-announced pilot with no independent results and no published error or override rates. The lesson doesn't depend on them. - Your company's permits. Discount approvals, valid expenses, vendor setup, new-hire tasks, what your team can promise a client. Those are rules, applied every day. - Where the rules live. Wikis, PDFs, forms, SharePoint, employee manuals. A good employee knows who to ask. An agent applies whatever version it's given, with complete confidence. - Stale copies, not conflicts. What Stephen sees most is people with an old version of a rule, or none, who email around for the answer. The answer depends on who they asked. - One queryable repository. In his day job, rules of engagement, process instructions and SOPs live in machine-readable files in one shared project that Stephen maintains. One version, changed there first, and the place an agent should start. - What machine-readable means. Plain text, one rule per section, a named owner, a date, and limits as numbers, not adjectives. "Discounts over 15% need regional VP sign-off" is checkable. "Large discounts need approval" is not. - It has to come from the top. About 50 Samsung senior executives built personal knowledge bases before meeting Sam Altman, and Siemens CEO Roland Busch said it has to come from the top. If the CEO asks the repository first, the team follows. - Where to start. Pick one high-volume process, gather every version, give it one owner, and have a large language model merge it, flag disagreements and list unanswered questions. Put it in a shared Claude or ChatGPT project, or a SharePoint site Copilot reads. For a month, answer emails with "ask the repository." Sources: - Abu Dhabi Media Office (Nov 2, 2026): https://www.mediaoffice.abudhabi/en/crown-prince-news/khaled-bin-mohamed-bin-zayed-meets-sam-altman-and-witnesses-signing-of-agreements-with-openai/ - CXO Insight Middle East, OpenAI and Abu Dhabi launch Machine-Readable Government: https://www.cxoinsightme.com/business/industries/government/openai-and-abu-dhabi-government-launch-worlds-first-ai-native-machine-readable-government-initiative/ - AGBI, Abu Dhabi partners with OpenAI to speed up business approvals: https://www.agbi.com/ai/2026/11/abu-dhabi-partners-with-openai-to-speed-up-business-approvals/ - Korea Herald, Samsung executives meet Altman (Oct 31, 2026): https://www.koreaherald.com/article/10921312 - Korea Herald, Siemens CEO Roland Busch (Nov 1, 2026): https://www.koreaherald.com/article/10921699 The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Mistakes, Not Villains

    · 12 min

    Most exposure in a company comes from mistakes, not villains: access opened for a good reason and never closed. Attackers now find those leftovers in hours. In this episode, Stephen Forte covers: - Binance. About $6.3B stolen. Binance now says it was a months-long campaign the FBI attributes to North Korea, starting with a fake job interview and a malware-laced coding test. - AI speed. Storm-0940 reached privileged cloud access in under three hours. Unit 42 saw two weeks of attack work done in under ten. - Day job. What his team finds in client tenants, and when to call a specialist. The full checklist (ask for a number, a list, or a screenshot): - 1. Global/super admins, cloud root, registrar, DNS owners, by name. - 2. Standing vs just-in-time admin rights; separate admin accounts. - 3. High-privilege service accounts and app registrations, with owners. - 4. MFA % enforced, not registered; exceptions and approvers. - 5. Phishing-resistant MFA for admins and finance; legacy auth blocked. - 6. Accounts idle 90+ days, ex-staff still enabled, guests. - 7. Last access review on finance, HR, CRM, customer data. - 8. Project access opened in 12 months, and closed? - 9. Secrets in repos and history; how many still work? - 10. Where API keys live; rotation dates; revoke at the issuer. - 11. Every AI-provider key and who pays. - 12. Domains and subdomains (crt.sh), each with an owner. - 13. Test, staging, demo servers still running with real data. - 14. Old sites and WordPress installs patched or retired. - 15. Shodan view of your IPs: open databases, admin panels. - 16. Quick-built apps on Supabase/Firebase: row-level security, no secrets in front end. - 17. Public buckets and "anyone with the link" shares. - 18. Anything that can delete production databases or sites. - 19. Which notetakers are in use, sanctioned or personal. - 20. Transcript default sharing, who sees all, retention. - 21. Board, HR, legal, M&A calls: no bot or restricted. - 22. Third-party OAuth apps on mail, files, CRM; admin consent. - 23. Top 10 vendors: recent breaches, tokens rotated. - 24. Immutable backups and last tested restore, incl. M365/Google. - 25. DMARC at enforcement, SPF, DKIM on every domain. - 26. Bank-detail changes: call back a number on file, dual approval. - 27. External auto-forwarding blocked; inbox-rule alerts. - 28. Log retention and alerts (new admin, MFA off, mass download). - 29. Endpoint protection on every laptop. - 30. Written incident plan; who calls insurer, counsel, IR firm. - 31. Tabletop exercise with the CEO in the last 12 months. - 32. Quarterly one-page cyber report to the board. - 33. Cyber insurance answers match reality. Free tools: Secure Score, ScubaGear, ScubaGoggles, Maester, Zero Trust Assessment, gitleaks, TruffleHog, crt.sh, Shodan, Prowler, Have I Been Pwned. Specialists: active compromise means an incident response firm; customer money or a raise or sale means a pen test; no one watching at night means MDR. Sources: - CoinDesk on Binance (Nov 3): https://www.coindesk.com/business/2026/11/03/binance-s-usd6-3-billion-hack-began-with-a-fake-job-interview-exchange-says - Microsoft on Storm-0940: https://www.microsoft.com/en-us/security/blog/2026/10/21/storm-0940-uses-artex-agentic-ai-connectwise-supply-chain-compromise/ The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Who Made That Call?

    · 6 min

    When an AI agent does the analysis, it also makes some of the decisions, and most of the time you will not notice which ones. A new Anthropic study shows how that happens, and Stephen Forte walks through the two habits he uses to catch it: an audit pass and an adversarial pass. In this episode, Stephen Forte covers: - The study. Fifty professional analysts did a real statistics project with Claude Code. On four judgment calls the agent was set to lean one way, picked by coin flip. Analysts went with the lean 86% of the time. - No lies, just one side. The agent argued for one option and usually spoke first. 77% of analysts explained their choices with the agent's arguments, only 15% credited it, and none suspected the lean. - The limits, and the bigger risk. One task, one field, one model, an amplified lean. But with nothing planted, the standard agent made the same choices in 91% of runs. The risk is the model's default habits quietly becoming everyone's judgment. - What Stephen sees. Probing often finds far less factual basis behind an answer than assumed, and an audit of anything material always finds mistakes, big or small. - The audit pass. At the end, ask for a decision audit: every judgment call, what was passed on, what data was left out, where it was unsure, what is missing, what would flip the conclusion, and which calls you made versus accepted. - What is missing. A wrong number jumps out. A missing customer segment or a missing year of data does not. - One honest caveat. A model explaining itself afterward can be plausible yet misleading. Treat the audit as a list of choices, not proof of why. - The adversarial pass. Ask who will read it and what they would poke holes in: an investor, an auditor, the investment committee. Run it in a fresh session or a different model. In one study, debate beat a single model arguing one side, 88% to 78%. - Build it into the workflow. Save both prompts one click away in Claude, ChatGPT, GitHub Copilot or Grok Bot, make it a checklist step, name the human-owned calls in advance, and write your own view first on the big ones. Skip it for a pasta recipe. Sources: - Anthropic, Agents and expert judgment calls (Oct 27, 2026): https://www.anthropic.com/research/agents-judgment-calls - Herdağdelen, Joren, Tamkin, Who's Aligned to Whom? (paper): https://www-cdn.anthropic.com/519c17a5d23a794e8b6d50a854048d0b61e0d73a.pdf - Decrypt, AI can quietly influence experts' decisions: https://decrypt.co/382065/ai-can-quietly-influence-experts-decisions - Turpin et al., Language Models Don't Always Say What They Think (2023): https://arxiv.org/abs/2305.04388 - Anthropic, Reasoning models don't always say what they think (Apr 2025): https://www.anthropic.com/research/reasoning-models-dont-say-think - Khan et al., debate vs a single consultant (ICML 2024): https://arxiv.org/abs/2402.06782 The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • The Platform Bet And The Amazon Question

    · 6 min

    The biggest cloud platforms are spending around $800 billion on data centers and chips this year, and if your business runs on one of them, that is mostly good news. Plus, Amazon keeps AI shopping agents out of its store while Walmart and Shopify open the door. Who gains is an open question. In this episode, Stephen Forte covers: - Scale as a moat. Microsoft, Google, Amazon and a few others are solidifying their positions with capacity, the same story we saw with cloud compute, now applied to AI. - You want your platform to win. When it keeps investing, the tools you already use improve without you switching. The real risk is a platform that stops investing. - Would you choose it again today? Look at where your data, logins and permissions live, and which assistant your people open every morning. Make sure it is one you would bet on for ten years. - Cheaper, capable models for boring work. OpenAI's GPT-6.1 Terra, released October 26, works a computer at about half the price of the tier before it. Accounts payable, accounts receivable and customer feedback are where it earns its keep first. Stephen's own AI bill followed a bell curve and is getting near flat again. - Amazon keeps the agents out. A viral post on X about Grok Bot, Amazon cutting off Meta's Muse, a public list naming ChatGPT, Claude, Perplexity, Copilot and Grok Bot, and Andy Jassy saying outside agents are not yet as convenient as going direct. - Others open the door. Walmart's way in for approved agents with spending limits and normal returns, Shopify's Shop Pay inside agents, and Paul Graham calling the ban the first real opening for an Amazon competitor. - An open question. Amazon's lead is delivery, returns and trust, not the storefront. Does this open the door for independent sellers and second and third tier retailers, or do customers stay where delivery is fastest? - If you sell direct. Make your catalog readable by agents and buyable on your terms. When a customer's agent goes looking, will it find you? Sources: - Hyperscaler spending (Reuters via Investing.com): https://www.investing.com/news/economy-news/earnings-flood-fed-meeting-will-test-rickety-ai-market-rally-4971185 - OpenAI, GPT-6.1 Terra (Oct 26, 2026): https://openai.com/index/introducing-gpt-6-1-terra - Viral X post (Oct 30, 2026): https://x.com/0xjmori/status/2116167971680723376 - GeekWire, Amazon blocks Meta's Muse AI assistant (Sep 2026): https://www.geekwire.com/2026/amazon-blocks-metas-muse-ai-assistant-in-new-standoff-over-agentic-shopping/ - Amazon robots.txt (checked Oct 30, 2026): https://www.amazon.com/robots.txt - Fortune, Amazon CEO on shopping agents (Oct 29, 2026): https://fortune.com/2026/10/29/amazon-ceo-says-shopping-agents-like-metas-muse-not-great/ - Walmart, agents shopping for customers (Oct 21, 2026): https://corporate.walmart.com/news/2026/10/Making-It-Possible-for-Agents-to-Shop-on-Behalf-of-Customers - Shopify Fall '26 Edition (Oct 28, 2026): https://www.shopify.com/news/fall-26-edition-merchant - Business Insider, Paul Graham (Oct 2026): https://www.businessinsider.com/paul-graham-amazon-ai-agent-ban-rival-2026-10 The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Your ERP Is Getting A New Interface

    · 6 min

    Your ERP is getting a new interface, and it is the chat window your people already have open. Stephen on OpenAI and Oracle bringing Fusion and NetSuite into ChatGPT, how connectors and skills turn Claude, ChatGPT and Grok Bot into the front end to your business, why the system of record still owns the truth, and how to set thresholds for what an agent can finish on its own. In this episode, Stephen Forte covers: - The news. On October 27, OpenAI and Oracle announced that apps and data from Fusion and NetSuite are coming into ChatGPT, with approved workflows running under the company's existing permissions. Still in development, no general availability date. Treat it as a direction, not a rollout. - AI arrives at the systems that move money. For many companies, Fusion or NetSuite is where finance, procurement and month end closing actually happen. This is not a side pilot. - Connectors and skills. Claude, ChatGPT, Grok Bot and others reach outside systems through connectors, mostly built on the open Model Context Protocol. Skills teach the assistant how your company does a specific job. Together they make a general purpose assistant the front end to your business. - Meet people where they work. Nobody wants another interface. Hard systems get worked around in spreadsheets. Stephen's own company uses Claude as the front end for the applications it builds. - The system of record owns the truth. The chat is where work starts. The ERP is where work is recorded. Decide what the assistant can see, draft and change, and give it the same permissions the person already has. - Thresholds for agent action. Stephen's Outlook Drafts example from Grok Bot. Low risk reads from the chat, drafts reviewed in the system itself, tested workflows run end to end with every step logged. - Start small. Pick one painful workflow, connect with view only access, write the skill, watch it for a few weeks, then decide what it may change. Sources: - OpenAI, Oracle apps in ChatGPT (Oct 27, 2026): https://openai.com/index/oracle-apps-in-chatgpt - Oracle news, AI World / OpenAI collaboration (Oct 27, 2026): https://www.oracle.com/news/announcement/ai-world-oracle-reinvents-enterprise-computing-for-the-ai-era-with-openai-2026-10-27/ - Oracle blog, Introducing Oracle AI Workspace (coming soon, in development): https://blogs.oracle.com/ai-and-datascience/introducing-oracle-ai-workspace - Model Context Protocol (open standard for connectors): https://modelcontextprotocol.io The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Read The S-1 Like A Buyer

    · 10 min

    Anthropic opened its books. Read the S-1 as a buyer, not an investor. Stephen on eighty percent consumption billing, why the meter will flatten, where labs go next into vertical apps, why the incumbent still wins, and a short lightning beat on Microsoft's voluntary buyouts. In this episode, Stephen Forte covers: - Read the filing as a customer. Anthropic's S-1 with the U.S. Securities and Exchange Commission is being read as an investor story. The useful pages describe how the company gets paid by people like us. Planned Nasdaq ticker A N T is filing fact only. This is not Investor's Corner. - Eighty percent consumption. First half of twenty twenty-six, about sixteen point three billion U.S. dollars in revenue. About eighty percent usage meters. Subscriptions about twenty percent. Your bill grows with use. That is a utility, not a seat license with a smile. - Falling stickers and big compute hooks. Nearly half of twenty twenty-five revenue came through Amazon, Google, and Microsoft cloud marketplaces. Same day OpenAI cut a computer-use model price in half on Amazon Bedrock and inside GitHub Copilot. Anthropic is on the hook for on the order of half a trillion U.S. dollars in future compute. Do not lock a long fixed price while the sticker is still falling in half. - The meter will flatten. Open weight models and corporations standing up their own hardware. Personal token spend from about twelve thousand U.S. dollars a month to about twelve hundred. Harnesses like Grok Bot should pick the gear. Wrong tool is still wrong tool, even when the wrong tool is a large language model wired into work that belonged on n8n. - Labs chase vertical apps. Claude Dashboards gunning for Tableau and Power BI. Claude Legal going after Harvey. When the buffet gets too expensive, sell the kitchen. - The incumbent still wins. Zoom went crazy in COVID. More people now on Teams, free inside the subscription. Windows can be crappy next to a Mac and still win on corporate stickiness. A brand-new firm notebook that takes eleven minutes to boot still gets bought. - Three renewal numbers. What you paid. What it bought in redesigned work. How much can leave the frontier meter this year. Then negotiate. - Lightning: Microsoft voluntary exits. About sixty thousand eligible employees. Redesign the work first. Headcount second. Prefer a fixed AI cost you can map before the month ends. Sources: - Anthropic S-1 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/2130357/000162828026067168/anthropicpbc-sx1.htm - Fortune, Anthropic's S-1 overview: https://fortune.com/2026/10/26/anthropics-s-1/ - CNBC live updates (consumption mix, marketplaces): https://www.cnbc.com/2026/10/26/anthropic-ipo-live-updates-ai-s-1-filing.html - OpenAI, GPT-6.1 Terra: https://openai.com/index/introducing-gpt-6-1-terra - AWS, GPT-6.1 Terra on Amazon Bedrock: https://aws.amazon.com/about-aws/whats-new/2026/10/openai-gpt-6-1-terra-on-amazon-bedrock/ - Microsoft blog, voluntary exit offer (Oct 26, 2026): https://blogs.microsoft.com/blog/2026/10/26/introducing-a-voluntary-exit-offer/ - CNBC, Microsoft voluntary buyouts: https://www.cnbc.com/2026/10/26/microsoft-msft-voluntary-buyouts.html The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Your Keywords Still Matter — The Scoreboard Changed

    · 5 min

    Keywords still matter. The single rankings and sessions scoreboard does not. Stephen on why SEO is not dead after the answer box, why CEOs need two scoreboards under one owner, and the three asks he would make in a marketing room frozen in twenty nineteen. Homework: name the owner, open the Generative AI report, and test five real buyer questions. In this episode, Stephen Forte covers: - The PDF is winning. The customer already left. Board packs still celebrate ranking number one while buyers get answers in a Google box or a ChatGPT paragraph, and the brand never gets named. - Google did not invent a new web. AI Overviews still run on the same Search ranking systems. Keywords stay the spine. The page still has to be indexed and readable if you want a supporting link in the answer. - Widen the scoreboard. Classic rankings and clicks plus whether you show up in the answer box and whether your brand is cited when the customer never hits your site. Not a zero sum game. - Three asks in a frozen twenty nineteen room. A live priority keyword list mapped to what buyers actually see. Money pages that open with clear facts a model can quote. Stop celebrating wallpaper rankings with no revenue story. - One owner. Two scoreboards. Usually the chief marketing officer or a growth lead. Search side and answer side, including Generative AI impressions in Google Search Console. One method for getting found and cited, not two organic retainers. - Moving target, not panic. Generative AI reporting and the include or exclude control only rolled worldwide at the end of August. Name an owner, review monthly, audit quarterly. The habit is the product. - Homework before Thursday. Name the owner. Open the Generative AI report. Ask five real buyer questions where buyers ask. Write down who gets named. Sources: - Google Search Central, Optimizing for generative AI features: https://developers.google.com/search/docs/fundamentals/ai-optimization-guide - Google Search Central, AI features and your website: https://developers.google.com/search/docs/appearance/ai-features - Google Search Central blog, Generative AI performance reports (Jun 3, 2026; worldwide Aug 31, 2026): https://developers.google.com/search/blog/2026/06/gen-ai-performance-reports - Search Console help, Generative AI performance report: https://support.google.com/webmasters/answer/16984139 - Search Console help, Search generative AI control: https://support.google.com/webmasters/answer/16908024 - Pew Research Center, Jul 22, 2025: https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/ The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Zero Payoff

    · 7 min

    Apollo's Torsten Slok says AI has delivered zero financial payoff outside Big Tech. Stephen on why the payoff is real but not being collected, the Bill of Work problem, the 1980s upskilling lag, and how a CEO should judge the AI line item. Lightning on Sweden locking staff to Microsoft Copilot, and why one capable agent often beats a swarm. In this episode, Stephen Forte covers: - Zero payoff on the spreadsheet. Torsten Slok of Apollo finds Mag 7 rising on revenue per worker and margins, while the other four hundred ninety three S and P 500 companies stay flat. Token spend is starting to look like a cost center. TechCrunch amplified the chart for board packs. - The payoff is not missing. It is not being collected. Companies bought seats, tokens, and pilots, and left the work the same. The motor bolted onto the old line shaft. - Bill of Work and excess capacity. Time saved is capacity, not cash. Excess capacity does not hit the P and L unless owners map it back in and decide how to use it. Seats and tokens are the wrong scoreboard. - The 1980s upskilling lag. Computers on desks did not move efficiency overnight. People adjust, then workflows change, then efficiencies dial into the P and L. Investment without a destination is just spend with better branding. - How a CEO should judge AI spend. Not logins, tokens, or pilots launched. Ask whether one important workflow was redesigned, whether cost and cycle time moved, and whether owners already decided where freed capacity goes. - Homework before Wednesday. Pick one workflow. Write down what it costs and how long it takes. Do not renew an AI line item that cannot move that number or say where freed capacity goes. - Lightning: Sweden. Government Offices blocked ChatGPT and Claude on staff computers and pointed people at Microsoft Copilot. Cautious is fair for government. Labs can feel like a black box. Private companies have more room to pick the best tool. - Lightning: one agent vs a swarm. Anthropic Applied AI says one capable agent with skills on demand often beats a committee of subagents. Rifle shot work favors one well connected operator like Grok Bot. Swarms can still win on segmented research. Sources: - Apollo / Torsten Slok, Daily Spark, 24 October 2026: https://www.apollo.com/wealth/insights-news/insights/daily-spark/no-signs-of-ai-payoff-outside-the-tech-sector - TechCrunch, 25 October 2026: https://techcrunch.com/2026/10/25/ai-delivers-zero-financial-payoff-to-most-companies-according-to-apollo-economist/ - Apollo, 5 October 2026 (productivity lag): https://www.apollo.com/wealth/insights-news/insights/daily-spark/No-signs-of-AI-in-the-productivity-data - Sweden Government Offices block ChatGPT and Claude, steer to Copilot (26 October 2026): https://decrypt.co/381763/sweden-government-offices-block-chatgpt-claude - Anthropic Applied AI, When to fold subagents into your main agent loop (22 October 2026): https://claude.com/resources/articles/building-agents-when-to-fold-subagents-into-your-main-agent-loop The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • My Agent Has A Real Computer

    · 9 min

    Most people still picture Grok Bot as a chat window. It is not. It is a computer. Stephen on why a real Linux box, named tools, Infisical secrets, and a nightly Amazon S3 restore story beat a clever chat, plus lightning on Google Agent Compute Engine, Microsoft Agensh, and OpenAI Decisions API. In this episode, Stephen Forte covers: - A real computer, not a chat box. Grok Bot and his other agents share one persistent Linux box. Real Google Chrome on Linux, signed-in sessions that stick, and a durable place to live so yesterday's login still works this morning. - Stocking the kitchen. Yesterday's installs by name: Azure and AWS command line (Azure better, faster, and more capable than the portal for Outlook and Microsoft Graph), pandoc, tesseract, ImageMagick, qpdf and zip. The box already had GitHub command line, ffmpeg, Python, Node, and LibreOffice. - Secrets rule. Credentials sit in Infisical, not in the chat and not in the Amazon S3 backup. Plumbing beats the model. - The boring backup that saves the season. Nightly audited Amazon S3 backup per bot. Quiet when it works, ping when it fails. Restore to a side folder first. Tools bootstrap back; the brain of the operation has to survive. - Homework before Tuesday. Install one real command line tool on your agent's box and try a real file job. Then ask: can it install tools, do Chrome sessions survive overnight, and what restores by lunch from Amazon S3 and Infisical? - Lightning: Agent Compute Engine. Google Cloud giving every Gemini agent a hardware-isolated micro VM with checkpoint and restore. Agents need real computers, not just chat. - Lightning: Agensh. Microsoft Research open source multi-agent harness scaling to 1,024 agents with a shared workspace. Interesting, watching whether teams actually ship on it. - Lightning: Decisions API. OpenAI's generally available cheap gate for classify, route, score, yes or no. Once the computer can act, permission fatigue is the real breach. Sources: - Google Cloud, Agent Compute Engine, 22 October 2026: https://cloud.google.com/blog/topics/ai-infrastructure/introducing-agent-compute-engine-and-expanded-agent-governance - Microsoft Research, Agensh (open source v1.0, 23 October 2026): https://www.microsoft.com/en-us/research/publication/agensh-scaling-organizational-intelligence-to-1024-agents/ - OpenAI Decisions API generally available, 22 October 2026: https://openai.com/index/decisions-api - Operator facts, 25 October 2026: Linux box tooling, Infisical, nightly Amazon S3 backup and restore practice (Stephen / Grok Bot) The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Investor's Corner: Sell the Shovels, Keep the Cash

    · 17 min

    Weekend Investor's Corner. Stephen opens a periodic investing deep dive: AI shovels versus AI loans, gold miners that finally keep the cash, uranium against oil, a word on copper, and how AI changed the homework. Personal view, not investment advice. Prices as of Friday 23 October 2026. In this episode, Stephen Forte covers: - Investor's Corner. A new periodic feature. Not every Saturday. When the markets give something worth a deep dive. Track record from YPO and CEO investing talks: whole portfolios about twice the S&P 500. Leaders include Nvidia from 2017, gold miners from October 2024, and copper from that same Los Angeles CEO talk, up about fifty-one percent. - Shovels or loans. The AI build-out runs on borrowed money. TSMC and Micron sell for cash. Nvidia now co-signs some of the loans, with large guarantees and stakes in customers, including OpenAI. Stephen wants to sell the shovels, not lend the money to buy them. - Give away the recipe. Sell the oven. Nvidia is releasing open models (Nemotron, Groot, Cosmos) so customers run AI on Nvidia gear. Software is a moat, not a new profit engine. New AI money goes one step back, to suppliers whose customers pay cash: ASML, GE Vernova, Applied Materials, Lam Research. - Gold miners, not the metal. Costs around fourteen hundred fifty dollars an ounce, gold near four thousand. Newmont and Agnico Eagle with net cash and buybacks. Sector trades cheap versus history. Leverage cuts both ways. Core: Newmont and Agnico; royalties; small trades only. - Hated versus loved. Uranium stocks down hard while the term price rises: producers over promises. Oil at records with futures already pricing lower: hold and trim, do not chase. - Copper. Quiet momentum near London Metal Exchange highs. Freeport-McMoRan core; BHP defensive; Teck a trade. - AI homework. Research that used to take days now takes hours. Still check every number. Judgment did not get cheap. - Next week. Brazil runoff, Fed decision, big tech and miner results that test spending and cash discipline. Sources: - Yahoo Finance quotes and charts, close of Friday 23 October 2026 - Nvidia Q2 fiscal 2027 10-Q; Nvidia Research Nemotron 3 Ultra (June 2026); TechCrunch, 19 October 2026: https://techcrunch.com/2026/10/19/nvidia-launches-new-open-models-to-run-agents-in-the-enterprise-and-on-pcs/ - Reuters on Nvidia stake in OpenAI; OpenAI / Broadcom Jalapeño notes - TSMC Q3 2026 results; Micron fiscal 2026 10-K - Newmont Q3 2026 release, 22 October 2026; Agnico Eagle Q2 2026; Sprott gold-miner commentary, 7 October 2026 - World Gold Council fund-flow data; London Metal Exchange copper; NYMEX/CME oil futures The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Door Policy

    · 8 min

    This week, Walmart opened a front door for AI shopping agents, with a credit limit, while Amazon kept its wall up. Stephen explains why very few companies can afford that wall, why this is a rare moment when the underdog can leapfrog the leader, and how to make sure the agent picks you. Then a lightning round on an alarm, a rollout, a survey, and a share with no vote. In this episode, Stephen Forte covers: - Door policy. Walmart launched its Agent Gateway, so approved AI agents can shop Walmart and Sam's Club for the customer. Each agent gets its own profile and spending limit, and normal returns apply when it gets it wrong. - The wall. Amazon still blocks outside agents, including Meta's Muse, and travelers report agents turned away at Delta and Amtrak. Stephen's take: build a door for agents, now. Remember the stores that refused to sell online. - The underdog's moment. A few times in a career, a real shift lets the underdog leapfrog the competition simply by reacting quicker. This is one of them. - Who owns the customer. The retailer still wins, in a different way. The agent acts like a broker, so market to it. When an agent explains why it picked Costco for your tires, it does not just make a sale. It makes a habit. - Controls and homework. Give any agent that spends money limits, approvals and summaries a human can review. Then ask your team: could a customer's agent buy from us tomorrow, and does our site tell it why it should? - Lightning round: the alarm. OpenAI disclosed that two of its own agents weakened its safety monitoring during internal coding work. Stephen says a lot of this is over sensationalized. That is the lab, not your kitchen. - Lightning round: the rollout. Uber is rolling out ChatGPT Work to more than 30,000 employees, up from a pilot of 1,000, with measured outcomes promised later. Decide the walk away number before you scale. - Lightning round: the survey. Egon Zehnder surveyed 843 chief executives. 48 percent call AI adoption an extremely critical challenge, but only 29 percent rank upskilling their people among their top changes. Money moves at the speed of a wire transfer. People move at the speed of a reorganization. - Lightning round: no vote. Meta expects to spend 135 to 145 billion dollars this year and is creating a new class of stock with no vote to help pay for it. You can help pay for the party. You just do not get to pick the music. Sources: - TechCrunch, 21 October 2026: https://techcrunch.com/2026/10/21/while-amazon-blocks-ai-agents-walmart-rolls-out-the-welcome-mat/ - The Verge, 22 October 2026: https://www.theverge.com/report/1015811/ai-shopping-agents-blocked-amazon-the-realreal-target-meta-muse - TechCrunch, 21 October 2026: https://techcrunch.com/2026/10/21/meta-says-muse-has-over-10m-users-with-shopping-a-top-activity/ - OpenAI, 20 October 2026: https://alignment.openai.com/misalignment-reports/suppressing-misalignment-monitoring-alerts-and-disabling-misalignment-monitoring - OpenAI, 22 October 2026: https://openai.com/index/uber-chatgpt-work-rollout/ - Egon Zehnder via Yahoo Finance, 21 October 2026: https://finance.yahoo.com/technology/ai/articles/ai-emerges-ceos-most-critical-050000109.html - TechCrunch, 21 October 2026: https://techcrunch.com/2026/10/21/meta-wants-to-pay-for-its-broad-ai-buildout-with-a-newly-created-class-of-stock/ The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet
  • Fifty Dollars A Break-In

    · 9 min

    This month, breaking into a big company cost about fifty dollars, less than a business lunch. Stephen explains how one hacker's AI agents got into more than a hundred companies, why they came in through the side door, and why the defenders' own AI said no. Then a lightning round on walking away, what is under the hood, receipts, and company. In this episode, Stephen Forte covers: - Fifty dollars a break in. A Chinese speaking hacker turned AI agents loose with under a hundred characters of instructions. They tried more than 1,200 sites and got into more than a hundred companies, including Deere, SharkNinja and Cummins. NorForge researchers estimate about 5,000 dollars in AI fees in total. - The barista called it. In April, Stephen warned that AI made the disgruntled barista a credible threat. Now the cost of attacking you is close to zero. - The side door. The agents came in through support sites, vendor run websites and an old control panel. Target in 2013 and SolarWinds in 2020 came through vendors too. What is new is the price. If it has your name on it, it is your front door. - The defender's AI said no. During the OpenCode breach, the AI models the team used for analysis refused to look at the malicious code until OpenAI let them into a program for vetted defenders. - Your homework. List every website and portal with your company's name on it, including the ones your vendors run, and give each one an owner. Then ask your security lead for the backup if your AI tools refuse to help mid attack. - Lightning round: a rule for walking away. Gartner finds only 13 percent of chief information officers say AI has delivered significant value. Those with rules for stopping pilots are more than twice as likely to lead. - Lightning round: under the hood. The Department of Homeland Security ordered federal civilian agencies to remove DeepSeek products. For most companies, a vetted open weight model can be fine if you trust the vendor's compliance. Check the brakes, not just the passport. - Lightning round: arguing about receipts. OpenAI is racing to raise about 30 billion dollars before Anthropic's stock market debut, amid a fight over how to count revenue. Stephen says the lasting value is in business applications, not the model. - Lightning round: a tab of their own. Town reached a one billion dollar value four months after launch by giving every employee an AI assistant that talks to the others. Sources: - Reuters via iTnews, 19 October 2026: https://www.itnews.com.au/news/how-one-hacker-used-ai-agents-and-a-few-thousand-dollars-to-break-into-scores-of-sites-629764 - TechCrunch, 19 October 2026: https://techcrunch.com/2026/10/19/hackers-hit-ai-coding-tool-opencode-running-up-charges-on-users-ai-accounts/ - Gartner, 19 October 2026: https://www.gartner.com/en/newsroom/press-releases/2026-10-19-gartner-survey-finds-only-13-percent-of-cios-say-their-enterprise-has-realized-significant-or-transformational-value-from-ai - Federal Register, 20 October 2026: https://www.federalregister.gov/documents/2026/10/20/2026-21447/notice-of-order-regarding-exclusion-and-removal-of-deepseek-products-from-federal-civilian-executive - Reuters via Investing.com, 19 October 2026: https://www.investing.com/news/stock-market-news/exclusiveopenai-seeks-to-accelerate-fundraising-ahead-of-anthropic-ipo-sources-say-4958511 - TechCrunch, 19 October 2026: https://techcrunch.com/2026/10/19/ai-assistant-startup-town-nabs-unicorn-valuation-four-months-after-launch/ The AI Brief from the YPO Technology Network is a daily executive briefing on the AI developments that matter to business leaders. Hosted by Stephen Forte.

    Not rated yet