{
  "apiVersion": "v1",
  "topic": {
    "id": "controlling-ai-spend",
    "title": "How do organizations keep AI spending in check?",
    "group": "Results and cost",
    "summary": "Caps, budgets, cheaper default models and routing, and the organizations that chose not to limit spending.",
    "passages": 17,
    "url": "https://the-ai-record.com/topics/controlling-ai-spend",
    "json": "https://the-ai-record.com/api/v1/topics/controlling-ai-spend.json"
  },
  "passages": [
    {
      "id": "notion-launched-workers-cpus-over-gpus-token-poor-2026",
      "text": "And the last thing is CPUs over GPUs. We've recently launched something at Notion called Workers. I don't think that the GPU is necessary for every job. A lot of the jobs that we have are actually serving discrete pieces of code. Like you don't need an LLM to turn a CSV into a PDF. You don't need an LLM to talk to Notion tool calls if we have a CLI. You definitely don't need an LLM to do deterministic SQL queries. This is where people become token poor very quick.",
      "context": "Sachs is closing a list of ways to lower AI costs.",
      "speaker": "Sarah Sachs",
      "speakerId": "sarah-sachs",
      "speakerUrl": "https://the-ai-record.com/people/sarah-sachs",
      "speakerLevel": "leader",
      "role": "Leads engineering teams for AI, Notion",
      "voice": "person",
      "organization": "Notion",
      "organizationId": "notion",
      "organizationUrl": "https://the-ai-record.com/organizations/notion",
      "aiVendor": true,
      "work": {
        "title": "Notion's Token Town",
        "type": "talk",
        "date": "2026-07-23",
        "publisher": "AI Engineer (YouTube)",
        "locator": "16:57",
        "url": "https://www.youtube.com/watch?v=-I5W5QVAT8E"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "where-agents-run",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering",
        "product"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/notion-launched-workers-cpus-over-gpus-token-poor-2026",
      "json": "https://the-ai-record.com/api/v1/passages/notion-launched-workers-cpus-over-gpus-token-poor-2026.json"
    },
    {
      "id": "notion-not-all-traffic-equal-triage-email-opus-2026",
      "text": "And not all traffic is equal. It is a huge miss to send all of these to the latest Opus model. Some of these absolutely. Large scale data analysis, when you do it on Notion, will recommend Opus, right? When you triage an email inbox, if we're charging you to do that on Opus, we're ripping you off, and ourselves.",
      "context": "Sachs is arguing that an AI product should route different kinds of requests to different models.",
      "speaker": "Sarah Sachs",
      "speakerId": "sarah-sachs",
      "speakerUrl": "https://the-ai-record.com/people/sarah-sachs",
      "speakerLevel": "leader",
      "role": "Leads engineering teams for AI, Notion",
      "voice": "person",
      "organization": "Notion",
      "organizationId": "notion",
      "organizationUrl": "https://the-ai-record.com/organizations/notion",
      "aiVendor": true,
      "work": {
        "title": "Notion's Token Town",
        "type": "talk",
        "date": "2026-07-23",
        "publisher": "AI Engineer (YouTube)",
        "locator": "10:24",
        "url": "https://www.youtube.com/watch?v=-I5W5QVAT8E"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "product-and-design",
        "controlling-ai-spend",
        "choosing-models"
      ],
      "functions": [
        "product",
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/notion-not-all-traffic-equal-triage-email-opus-2026",
      "json": "https://the-ai-record.com/api/v1/passages/notion-not-all-traffic-equal-triage-email-opus-2026.json"
    },
    {
      "id": "juspay-multi-model-strategy-cheaper-models-simple-checks-cut-costs-2026",
      "text": "Multi-model strategy works: Using cheaper models for simple checks and expensive ones for complex reasoning cut costs by 80%",
      "context": null,
      "speaker": null,
      "speakerId": null,
      "speakerUrl": null,
      "speakerLevel": null,
      "role": null,
      "voice": "organization",
      "organization": "Juspay",
      "organizationId": "juspay",
      "organizationUrl": "https://the-ai-record.com/organizations/juspay",
      "aiVendor": true,
      "work": {
        "title": "How We Built an AI Code Reviewer with NeuroLink and Bitbucket",
        "type": "post",
        "date": "2026-04-06",
        "publisher": "dev.to (NeuroLink AI organization account, Juspay)",
        "locator": "Results & Lessons Learned > Key Lessons",
        "url": "https://dev.to/neurolink/how-we-built-an-ai-code-reviewer-with-neurolink-and-bitbucket-3io9"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "code-review",
        "ai-review-cost",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [
        {
          "metric": "cost-reduction-multi-model",
          "value": 0.8,
          "unit": "share",
          "definition": "Cost cut by routing simple checks to cheaper models and complex reasoning to expensive ones",
          "level": "n/a"
        }
      ],
      "url": "https://the-ai-record.com/passages/juspay-multi-model-strategy-cheaper-models-simple-checks-cut-costs-2026",
      "json": "https://the-ai-record.com/api/v1/passages/juspay-multi-model-strategy-cheaper-models-simple-checks-cut-costs-2026.json"
    },
    {
      "id": "uber-spend-tiers-manager-approval-roi-conversation-2026",
      "text": "And we have this notion of spend tiers. So by default, you get certain … tier, and then we have many levels of tiers. Once you're close to that tier, then you can raise a request to go to the next tier. And you just need a simple manager approval. … But we don't want to restrict engineers … they should be able to spend whatever they need because ultimately we want to … maximize the gain from this.",
      "context": "Answering the host's question whether an engineer at Uber needs to be aware of AI spend.",
      "speaker": "Uday Kiran Medisetty",
      "speakerId": "uday-kiran-medisetty",
      "speakerUrl": "https://the-ai-record.com/people/uday-kiran-medisetty",
      "speakerLevel": "practitioner",
      "role": "Distinguished Engineer, Uber",
      "voice": "person",
      "organization": "Uber",
      "organizationId": "uber",
      "organizationUrl": "https://the-ai-record.com/organizations/uber",
      "aiVendor": false,
      "work": {
        "title": "Uber's AI infra, from the inside",
        "type": "podcast",
        "date": "2026-10-14",
        "publisher": "The Pragmatic Engineer (YouTube)",
        "locator": "26:54",
        "url": "https://www.youtube.com/watch?v=MDUYSMJmi8A"
      },
      "grade": "C",
      "gradeMeaning": "An interview or press quote.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "controlling-ai-spend",
        "ai-cost-and-spend"
      ],
      "functions": [
        "engineering",
        "finance"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/uber-spend-tiers-manager-approval-roi-conversation-2026",
      "json": "https://the-ai-record.com/api/v1/passages/uber-spend-tiers-manager-approval-roi-conversation-2026.json"
    },
    {
      "id": "mercado-libre-ramp-up-our-use-third-party-ai-2026",
      "text": "We expect this leverage to be partially offset by the ramp up in our use of third-party AI models. This reflects a deliberate choice to prioritize company-wide adoption over cost optimization at this stage. This investment is contributing to cost efficiency across many other expense lines (customer service, for example) as well as helping us to generate incremental revenue (in Acquiring and Advertising, for example).",
      "context": null,
      "speaker": null,
      "speakerId": null,
      "speakerUrl": null,
      "speakerLevel": null,
      "role": null,
      "voice": "organization",
      "organization": "Mercado Libre",
      "organizationId": "mercado-libre",
      "organizationUrl": "https://the-ai-record.com/organizations/mercado-libre",
      "aiVendor": false,
      "work": {
        "title": "MercadoLibre, Inc. Q1 2026 results (Form 8-K, Exhibit 99.1)",
        "type": "filing",
        "date": "2026-05-07",
        "publisher": "SEC EDGAR",
        "locator": "Costs and expenses, Product Development",
        "url": "https://www.sec.gov/Archives/edgar/data/1099590/000109959026000014/meli-20260507xex991.htm"
      },
      "grade": "A",
      "gradeMeaning": "A filing or earnings call the organization answers for.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "adoption-and-usage",
        "controlling-ai-spend",
        "support-cost"
      ],
      "functions": [
        "company-wide"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/mercado-libre-ramp-up-our-use-third-party-ai-2026",
      "json": "https://the-ai-record.com/api/v1/passages/mercado-libre-ramp-up-our-use-third-party-ai-2026.json"
    },
    {
      "id": "airbnb-glowc-fine-tuned-classifier-matches-frontier-quality-2026",
      "text": "GLOWC, an internal framework used by engineers at Airbnb, makes it easy to fine-tune open-weight language (and vision-language) models into fast, accurate classifiers. In this blog post, we put one of these classifiers head-to-head with open-weight and proprietary models and find that it matches frontier-level quality in under 100ms at roughly 1% of the cost.",
      "context": null,
      "speaker": null,
      "speakerId": null,
      "speakerUrl": null,
      "speakerLevel": null,
      "role": null,
      "voice": "organization",
      "organization": "Airbnb",
      "organizationId": "airbnb",
      "organizationUrl": "https://the-ai-record.com/organizations/airbnb",
      "aiVendor": false,
      "work": {
        "title": "Frontier quality at 1% of the cost: a production benchmark of LLM classifiers",
        "type": "post",
        "date": "2026-11-03",
        "publisher": "Airbnb Tech blog",
        "locator": "Introduction",
        "url": "https://airbnb.tech/ai-ml/frontier-quality-at-1-of-the-cost-a-production-benchmark-of-llm-classifiers/"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-05"
      },
      "topics": [
        "internal-tools",
        "ai-cost-and-spend",
        "controlling-ai-spend",
        "choosing-models"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [
        {
          "metric": "classifier-cost-vs-frontier",
          "value": 0.01,
          "unit": "share",
          "period": "2026 benchmark",
          "definition": "approximate cost of the fine-tuned classifier relative to frontier models on one production policy task, at public list rates and single-GPU rental cost",
          "level": "human-out"
        }
      ],
      "url": "https://the-ai-record.com/passages/airbnb-glowc-fine-tuned-classifier-matches-frontier-quality-2026",
      "json": "https://the-ai-record.com/api/v1/passages/airbnb-glowc-fine-tuned-classifier-matches-frontier-quality-2026.json"
    },
    {
      "id": "airbnb-proprietary-models-only-win-low-traffic-2026",
      "text": "Proprietary models only win at low traffic: below roughly 15 to 30 RPM, Claude Sonnet 5.5 and GPT-6.1 Sol are cheaper than a dedicated GPU.",
      "context": null,
      "speaker": null,
      "speakerId": null,
      "speakerUrl": null,
      "speakerLevel": null,
      "role": null,
      "voice": "organization",
      "organization": "Airbnb",
      "organizationId": "airbnb",
      "organizationUrl": "https://the-ai-record.com/organizations/airbnb",
      "aiVendor": false,
      "work": {
        "title": "Frontier quality at 1% of the cost: a production benchmark of LLM classifiers",
        "type": "post",
        "date": "2026-11-03",
        "publisher": "Airbnb Tech blog",
        "locator": "Cost at scale",
        "url": "https://airbnb.tech/ai-ml/frontier-quality-at-1-of-the-cost-a-production-benchmark-of-llm-classifiers/"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-05"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend",
        "choosing-models"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/airbnb-proprietary-models-only-win-low-traffic-2026",
      "json": "https://the-ai-record.com/api/v1/passages/airbnb-proprietary-models-only-win-low-traffic-2026.json"
    },
    {
      "id": "cloudflare-code-quality-high-effort-average-review-2026",
      "text": "With code quality at high effort, the average review now costs just $0.77, down 68% from the $2.43 it averaged over our two most expensive weeks before we started.",
      "context": null,
      "speaker": "Ryan Skidmore",
      "speakerId": "ryan-skidmore",
      "speakerUrl": "https://the-ai-record.com/people/ryan-skidmore",
      "speakerLevel": "practitioner",
      "role": "Author of Cloudflare's \"Orchestrating AI Code Review at scale\" blog post",
      "voice": "person",
      "organization": "Cloudflare",
      "organizationId": "cloudflare",
      "organizationUrl": "https://the-ai-record.com/organizations/cloudflare",
      "aiVendor": true,
      "work": {
        "title": "Cutting our inference bill by 68% by being obsessive about optimization",
        "type": "post",
        "date": "2026-10-19",
        "publisher": "Cloudflare blog",
        "url": "https://blog.cloudflare.com/obsessive-inference-optimization/"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "ai-review-cost",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [
        {
          "metric": "cost-per-review",
          "value": 0.77,
          "unit": "usd",
          "period": "after mid-September to early October 2026 changes",
          "definition": "Average cost of one AI code review, code quality reviewer at high effort",
          "level": "n/a"
        },
        {
          "metric": "cost-per-review-reduction",
          "value": 0.68,
          "unit": "share",
          "period": "versus the two most expensive weeks before the work",
          "definition": "Fall in average cost per AI code review (from $2.43)",
          "level": "n/a"
        }
      ],
      "url": "https://the-ai-record.com/passages/cloudflare-code-quality-high-effort-average-review-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cloudflare-code-quality-high-effort-average-review-2026.json"
    },
    {
      "id": "spacex-grok-bot-compact-while-cache-warm-2026",
      "text": "One trick that we do, which I think is really helpful, is while you're still within the warm cache period, you should probably do the compaction right now, because you have this long conversation … going to be much cheaper. So just go ahead, do the compaction, and now when the user comes back at 30 minutes or whatever, they're starting from this summary, and the summary now is going to be much cheaper for them to continue adding messages on.",
      "context": "Robinson is describing how his team keeps the cost of long agent conversations down by summarizing the context before the model provider's prompt cache expires.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "32:20",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-grok-bot-compact-while-cache-warm-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-grok-bot-compact-while-cache-warm-2026.json"
    },
    {
      "id": "xebia-running-several-agents-parallel-burns-tokens-2026",
      "text": "Running several agents in parallel burns tokens quickly, so set budgets and keep an eye on spending from day one.",
      "context": null,
      "speaker": "Noah Blauensteiner",
      "speakerId": "noah-blauensteiner",
      "speakerUrl": "https://the-ai-record.com/people/noah-blauensteiner",
      "speakerLevel": "practitioner",
      "role": "Author, Xebia articles",
      "voice": "person",
      "organization": "Xebia",
      "organizationId": "xebia",
      "organizationUrl": "https://the-ai-record.com/organizations/xebia",
      "aiVendor": true,
      "work": {
        "title": "Shining Some Light in Dark Agentic Software Factories",
        "type": "post",
        "date": "2026-08-12",
        "publisher": "Xebia articles",
        "locator": "What is an Agentic Software Factory?",
        "url": "https://xebia.com/articles/shining-some-light-in-dark-agentic-software-factories/"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-10-31"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/xebia-running-several-agents-parallel-burns-tokens-2026",
      "json": "https://the-ai-record.com/api/v1/passages/xebia-running-several-agents-parallel-burns-tokens-2026.json"
    },
    {
      "id": "notion-auto-model-handles-about-75-traffic-2026",
      "text": "Notion's auto model does this really well. We have state-of-the-art models available always. But we also have an auto model there at the top that handles about 75% of our traffic, right? We have the ability to switch between models and our product, and we also offer it to our customers … they have access to these models without vendor lock-in.",
      "context": "Sachs is describing how Notion avoids depending on a single AI model provider.",
      "speaker": "Sarah Sachs",
      "speakerId": "sarah-sachs",
      "speakerUrl": "https://the-ai-record.com/people/sarah-sachs",
      "speakerLevel": "leader",
      "role": "Leads engineering teams for AI, Notion",
      "voice": "person",
      "organization": "Notion",
      "organizationId": "notion",
      "organizationUrl": "https://the-ai-record.com/organizations/notion",
      "aiVendor": true,
      "work": {
        "title": "Notion's Token Town",
        "type": "talk",
        "date": "2026-07-23",
        "publisher": "AI Engineer (YouTube)",
        "locator": "13:15",
        "url": "https://www.youtube.com/watch?v=-I5W5QVAT8E"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "product-and-design",
        "controlling-ai-spend",
        "choosing-models"
      ],
      "functions": [
        "product",
        "engineering"
      ],
      "figures": [
        {
          "metric": "auto-model-traffic-share",
          "value": 75,
          "unit": "percent",
          "period": "stated in the 2026 talk",
          "definition": "Share of Notion's traffic that its auto model handles, as Sachs says it: 'about 75% of our traffic'.",
          "level": "unclear"
        }
      ],
      "url": "https://the-ai-record.com/passages/notion-auto-model-handles-about-75-traffic-2026",
      "json": "https://the-ai-record.com/api/v1/passages/notion-auto-model-handles-about-75-traffic-2026.json"
    },
    {
      "id": "notion-evals-early-access-exchange-instead-large-commits-2026",
      "text": "That expertise on use cases is also very valuable to Frontier Labs. We find that our evals and our Early Access Program partnerships actually help us a lot with Frontier Labs and it's something that we can exchange instead of extraordinary large commits. And I don't think the discount is ever worth the lost in optionality.",
      "context": "Sachs is describing what Notion offers AI model labs in place of large spending commitments; 'that expertise' refers to Notion's evaluations of models on its own use cases.",
      "speaker": "Sarah Sachs",
      "speakerId": "sarah-sachs",
      "speakerUrl": "https://the-ai-record.com/people/sarah-sachs",
      "speakerLevel": "leader",
      "role": "Leads engineering teams for AI, Notion",
      "voice": "person",
      "organization": "Notion",
      "organizationId": "notion",
      "organizationUrl": "https://the-ai-record.com/organizations/notion",
      "aiVendor": true,
      "work": {
        "title": "Notion's Token Town",
        "type": "talk",
        "date": "2026-07-23",
        "publisher": "AI Engineer (YouTube)",
        "locator": "14:39",
        "url": "https://www.youtube.com/watch?v=-I5W5QVAT8E"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering",
        "company-wide"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/notion-evals-early-access-exchange-instead-large-commits-2026",
      "json": "https://the-ai-record.com/api/v1/passages/notion-evals-early-access-exchange-instead-large-commits-2026.json"
    },
    {
      "id": "coinbase-91-employees-were-never-hitting-their-usage-2026",
      "text": "91% of our employees were never hitting their usage caps, so instead of lowering caps and driving up alerts, we're moving to cheaper defaults.",
      "context": null,
      "speaker": "Brian Armstrong",
      "speakerId": "brian-armstrong",
      "speakerUrl": "https://the-ai-record.com/people/brian-armstrong",
      "speakerLevel": "executive",
      "role": "CEO",
      "voice": "person",
      "organization": "Coinbase",
      "organizationId": "coinbase",
      "organizationUrl": "https://the-ai-record.com/organizations/coinbase",
      "aiVendor": false,
      "work": {
        "title": "X post by unnamed account",
        "type": "social",
        "date": "2026-06-27",
        "publisher": "X post by Brian Armstrong, CEO",
        "url": "https://api.fxtwitter.com/brian_armstrong/status/2070670644577280109"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-10-31"
      },
      "topics": [
        "ai-across-the-company",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/coinbase-91-employees-were-never-hitting-their-usage-2026",
      "json": "https://the-ai-record.com/api/v1/passages/coinbase-91-employees-were-never-hitting-their-usage-2026.json"
    },
    {
      "id": "uber-cost-ai-too-damn-high-costs-gone-up-6x-since-2024-2026",
      "text": "… speaking of costs, the cost of AI is too damn high. … since 2024, our costs have gone up at least 6x. … again, no question that it's had positive impact, but it's gone from something that I can self fund using my own budget to something that I need to ask permission from … the CFO.",
      "context": "Chadha is describing how the cost of AI tools at Uber has grown and what it changed about who approves the spending.",
      "speaker": "Anshu Chadha",
      "speakerId": "anshu-chadha",
      "speakerUrl": "https://the-ai-record.com/people/anshu-chadha",
      "speakerLevel": "leader",
      "role": "Director of Engineering, Dev Platform, Uber",
      "voice": "person",
      "organization": "Uber",
      "organizationId": "uber",
      "organizationUrl": "https://the-ai-record.com/organizations/uber",
      "aiVendor": false,
      "work": {
        "title": "Uber: Leading engineering through an agentic shift - The Pragmatic Summit",
        "type": "talk",
        "date": "2026-03-10",
        "publisher": "The Pragmatic Engineer (YouTube)",
        "locator": "35:56",
        "url": "https://www.youtube.com/watch?v=i1tZN41VKcE"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering",
        "finance"
      ],
      "figures": [
        {
          "metric": "ai-cost-growth",
          "value": 6,
          "unit": "x",
          "period": "since 2024, at least",
          "definition": "growth in Uber's AI costs (Chadha says 'our costs') since 2024",
          "level": "n/a"
        }
      ],
      "url": "https://the-ai-record.com/passages/uber-cost-ai-too-damn-high-costs-gone-up-6x-since-2024-2026",
      "json": "https://the-ai-record.com/api/v1/passages/uber-cost-ai-too-damn-high-costs-gone-up-6x-since-2024-2026.json"
    },
    {
      "id": "cisco-given-him-unlimited-tokens-because-he-s-going-2026",
      "text": "I’ve given him unlimited tokens because he’s going to refactor code… So I’m like, you can go, you can go nuts.",
      "context": null,
      "speaker": "Jeetu Patel",
      "speakerId": "jeetu-patel",
      "speakerUrl": "https://the-ai-record.com/people/jeetu-patel",
      "speakerLevel": "executive",
      "role": "President and Chief Product Officer",
      "voice": "person",
      "organization": "Cisco",
      "organizationId": "cisco",
      "organizationUrl": "https://the-ai-record.com/organizations/cisco",
      "aiVendor": true,
      "work": {
        "title": "Cisco chief found a 27-year-old ex-founder ‘buried away in the organization doing something unstrategic’—now he’s upgrading 98% of its code",
        "type": "press",
        "date": "2026-10-30",
        "publisher": "Fortune (Orianna Rosa Royle)",
        "url": "https://fortune.com/2026/10/30/cisco-president-jeetu-patel-gen-z-27-year-old-unlimited-ai-tokens-rewrite-legacy-code-entry-level-jobs/"
      },
      "grade": "C",
      "gradeMeaning": "An interview or press quote.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-04"
      },
      "topics": [
        "migrations",
        "ai-cost-and-spend",
        "leadership-and-mandates",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/cisco-given-him-unlimited-tokens-because-he-s-going-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cisco-given-him-unlimited-tokens-because-he-s-going-2026.json"
    },
    {
      "id": "nordic-corporate-bank-agent-costs-100-to-200-a-month-2026",
      "text": "And how much does this get cost? I don't really have a clear answer to that because he's sharing my GitHub subscription. So I think I use most of it. But I think maybe it's been costing like $100 to $200 a month maybe. And that's not, that's almost for free. But I guess that will increase now. And also we're not using the most expensive models. So that cost could go up a lot. But it's enabled us to do this.",
      "context": "Asked what the autonomous agent costs, the speaker gives a rough monthly range and says it may rise.",
      "speaker": "Hallstein Brøtan",
      "speakerId": "hallstein-brotan",
      "speakerUrl": "https://the-ai-record.com/people/hallstein-brotan",
      "speakerLevel": "practitioner",
      "role": "Senior consultant, Novanet; developer on Nordic Corporate Bank's system",
      "voice": "person",
      "organization": "Nordic Corporate Bank",
      "organizationId": "nordic-corporate-bank",
      "organizationUrl": "https://the-ai-record.com/organizations/nordic-corporate-bank",
      "aiVendor": false,
      "work": {
        "title": "How We Built an Autonomous Development Workflow at Nordic Corporate Bank",
        "type": "talk",
        "date": "2026-06-29",
        "publisher": "NDC Conferences (YouTube); talk recorded at NDC in Oslo",
        "locator": "50:39",
        "url": "https://www.youtube.com/watch?v=p8M8--12h1Y"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-08"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [
        {
          "metric": "coding-agent-monthly-cost-low",
          "value": 100,
          "unit": "usd",
          "period": "per month, as spoken",
          "definition": "the speaker's rough estimate of what the agent costs a month; it shares his GitHub subscription and he says he has no clear answer",
          "level": "n/a"
        },
        {
          "metric": "coding-agent-monthly-cost-high",
          "value": 200,
          "unit": "usd",
          "period": "per month, as spoken",
          "definition": "upper end of the same rough estimate",
          "level": "n/a"
        }
      ],
      "url": "https://the-ai-record.com/passages/nordic-corporate-bank-agent-costs-100-to-200-a-month-2026",
      "json": "https://the-ai-record.com/api/v1/passages/nordic-corporate-bank-agent-costs-100-to-200-a-month-2026.json"
    },
    {
      "id": "uber-every-model-request-goes-through-one-gateway-2026",
      "text": "And since all the requests go through one place, we could also add like audit logging, session traces, and a whole bunch of infrastructure around this. Every model request at Uber goes through this. So there's no model request that doesn't go through this. So that gives us like one single entry point for across all the models.",
      "context": "'This' is Uber's internal model gateway, which he has just been describing (one OpenAI-compatible or Anthropic-compatible API in front of models from all vendors, with an AI Guard safety layer).",
      "speaker": "Uday Kiran Medisetty",
      "speakerId": "uday-kiran-medisetty",
      "speakerUrl": "https://the-ai-record.com/people/uday-kiran-medisetty",
      "speakerLevel": "practitioner",
      "role": "Distinguished Engineer, Uber",
      "voice": "person",
      "organization": "Uber",
      "organizationId": "uber",
      "organizationUrl": "https://the-ai-record.com/organizations/uber",
      "aiVendor": false,
      "work": {
        "title": "Uber's AI infra, from the inside",
        "type": "podcast",
        "date": "2026-10-14",
        "publisher": "The Pragmatic Engineer (YouTube)",
        "locator": "24:22",
        "url": "https://www.youtube.com/watch?v=MDUYSMJmi8A"
      },
      "grade": "C",
      "gradeMeaning": "An interview or press quote.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "internal-tools",
        "choosing-models",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/uber-every-model-request-goes-through-one-gateway-2026",
      "json": "https://the-ai-record.com/api/v1/passages/uber-every-model-request-goes-through-one-gateway-2026.json"
    }
  ],
  "notice": [
    "Every passage is reproduced word for word from the cited source.",
    "Check the verification status: \"reported\" means not yet confirmed by an independent second reader.",
    "Organizations flagged aiVendor sell AI tools or services and have a reason to publish flattering numbers.",
    "An empty result means the archive holds nothing on it, not that nobody has said anything."
  ]
}