Take the day one card at a time

• In the keynote's research segment the host said "A few weeks ago we reached that goal." and "We now have an AI research intern." (23:52)
• Patwardhan, who introduced herself as "a research lead for post training", said OpenAI's models "have taken on over a third of day long research tasks with no intervention required" (26:27)
• The figure is OpenAI's own measurement of its own models, given as of July; before that, she said, they "failed the vast majority of times" on such tasks (26:27)

OpenAI published its DevDay 2026 keynote on its own YouTube channel on 29 September 2026. The video's description names Sam Altman, Romain Huet, Tejal Patwardhan and Holly Li as the presenters. Introducing the research segment, the host said "Last year we maded a bold prediction saying we would see the first AI research intern in the text year" (23:34), the automatic captions' rendering of what appears to be next year. He then said "A few weeks ago we reached that goal." (23:48) and "We now have an AI research intern." (23:52). Tejal Patwardhan, who introduced herself as "a research lead for post training" (24:17), gave what is OpenAI's own measurement of its own models: they failed most day-long research tasks before, and "as of July and E-our models have taken on over a third of day long research tasks with no intervention required" (26:27). "E-our" is the captions' rendering of a second month name that is not clear. Holly Li, presenting the dots agents, said of OpenAI's own engineers: "Our engineers have their dot fixing dozens of bugs every day." (15:34). Altman also said "Around 1.2 billion people use ChatGPT." (44:11) and, on the API, "We've still kept more than 99 percent reliability." (31:06).

Like0Love0Insightful0Surprising0Funny0Not useful0

• The White House published Executive Order 14434 on 29 September; in the video Trump said "We're going to be signing a document today at about 5:00" (01:25)
• Asked by an unnamed reporter about "no guard rails, no regulation of AI", Trump said "there's a belief that there should be tremendous self-regulation" (02:04)
• Trump said the executives will "work with local communities for the data centers in particular" (02:22) and "Otherwise they'll be forced to go overseas or other locations." (03:03)

The White House published a video titled "President Trump Participates in a Meeting and Luncheon on Super Intelligence" on its YouTube channel on 29 September 2026. The same day the White House published Executive Order 14434, Inaugurating the Era of Super Intelligence. In the video Trump said "We're going to be signing a document today at about 5:00" (01:25) and "we're going to be renaming it super intelligence, officially renaming it" (01:35). A reporter asked "The belief of everybody that you're meeting with today that no guard rails, no regulation of AI, super intelligence as you call it is necessary." (01:54), as the automatic captions give the question. Trump answered "Well, there's a belief that there should be tremendous self-regulation and we automatically have regulation with the Department of Justice, the FBI, all of that." (02:04). He said the executives will "work with local communities for the data centers in particular" (02:22) and warned "Otherwise they'll be forced to go overseas or other locations." (03:03).

Like0Love0Insightful0Surprising0Funny0Not useful0

• AWS is releasing monthly project spend limits to a limited number of customers; at the limit it pauses the project and keeps data, deleting it after 90 days without action.
• Google Cloud's Spend Caps, introduced 29 July and in public preview, apply to a single project and service and cover Gemini API, Agent Platform, Cloud Run and Cloud Run Functions.
• Google says caps on AI services trigger within minutes and are non-destructive, but fixed commitment fees such as Committed Use Discounts or Provisioned Throughput keep billing at their flat rate.

AWS and Google Cloud each describe a monthly limit that stops usage when it is reached, but neither is generally available: AWS says it is releasing its experience to a limited number of customers, and Google's spend caps are in public preview. On 16 September 2026 AWS said builders can set a monthly spend limit for a project, and that if the project's usage reaches the limit, "your project is paused for that month". AWS's documentation describes a spend limit as "the most you'll pay per month for a project", set per project, so some projects can have one and others none, up to 10 projects. The minimum is "the greater of $20 or a conservative estimate of your likely spend", and limits exclude credits and apply to pre-tax charges. AWS notifies you when costs reach 50%, 75% and 90% of the limit, or are on track to exceed it within 10 days. Three optional controls act earlier: about 7 days before the limit new resources stop launching, about 5 days before it idle resources are paused, and about 4 days before it the highest-cost active resources are paused, drawn from EC2, RDS, Lambda, Bedrock and SageMaker. When the limit is reached, "AWS pauses your project and stops all resources. Your data is preserved.", and raising the limit reactivates it; if nothing is done within 90 days, AWS permanently deletes the project's data. A spend limit needs a Paid Plan. On 29 July 2026 Google Cloud introduced Spend Caps on Budgets, which set a monthly cap on a specific service within a project and, once reached, restrict further cost-incurring usage for that service. Google says caps for AI services trigger within minutes of the threshold, are non-destructive, and that fixed commitment fees such as Committed Use Discounts or Provisioned Throughput keep billing at their flat rate. In Google's public preview a cap applies to a single project and service for a fixed monthly timeframe, covers Gemini API, Agent Platform, Cloud Run and Cloud Run Functions, and stays in place until someone lifts it in the Budgets UI with one click.

Like0Love0Insightful0Surprising0Funny0Not useful0