AI CEOs Sign Safety Accord, PCE Inflation Comes In Cool
The Rundown
September 30, 2026
Market update for September 30, 2026 Interview with Tech Journalist Kevin Roose (Spotify, YouTube) In today’s episode, Zaid covers: Cooler-than-expected PCE inflation and whether the bond market can finally calm down President Trump’s new AI safety accord with the biggest names in tech OpenAI’s...
Speakers Zaid Admani
Zaid Admani (0:00)
Public.com presents The Rundown, your daily market update in 10 minutes. My name is Zaid Admani, and today is Wednesday, September 30th. In today's episode, we'll break down the latest inflation report and why it's finally giving the bond market some relief. We'll also tell you what happened when President Trump brought some of the biggest names in AI to the White House yesterday, and also recap OpenAI's huge DevDay announcements. Then stick around to the end of the show to find out why flying home for the holidays could cost you more than it has in a decade. We got a great show for you today.
Let's go.
Stocks finished slightly lower on Tuesday. The S&P 500 fell by 0.2% while the NASDAQ dipped by 0.1%.
And I'm kind of sick of saying this by now, but the bond market was the headliner once again. The 10 year and 30 year treasury yield both ticked higher. Despite other macro factors that would typically bring down yields. Like for example, oil prices fell by more than 2% yesterday because Saudi Arabia restarted their critical East West pipeline. On top of that, New York Fed President John Williams gave a relatively dovish speech yesterday, saying the Fed has no need for urgency when it comes to raising rates again. Those were some pretty shocking comments. It kind of caught the market off guard. In fact, the odds of an October rate hike went from around 70% before his speech down to roughly 50% according to prediction markets, which are now available on public.com. And the odds of a rate hike went down again this morning, dropping to 35% after the PCE inflation report for the month of August came out. This report is the Fed's preferred inflation gauge, and it came in cooler than expected. Headline PCE rose just 0.3% month over month, which was right in line with estimates. But on a year over year basis, inflation was 3.4% versus the 3.7% that was expected. And then if you look at core PCE, which strips out food and energy, that rose 0.2% for the month and was up 3% year over year, which was also cooler than the 3.3% expected. So I gotta say that was a pretty encouraging inflation report. The market seems to love it. I mean, treasury yields are dipping a bit and stock futures are moving higher in the pre-market at the time of this recording. Now keep in mind, this PCE report from this morning was for the month of August. So it is a bit dated, but the market seems to be focused on the good news. And who knows, maybe the bond market will finally chill out and yields will pull back some more. We'll continue to keep an eye on all of that. So definitely get subscribed to the podcast if you haven't already and tune in every day to stay in the loop.
Let's run through some headlines.
And we're talking AI. Yesterday, President Trump hosted a lunch at the White House with some of the biggest names in AI. Nearly two dozen executives and AI leaders were there. I'm talking Jensen Huang, Elon Musk, Dario Amadei, Mark Zuckerberg, Sundar Pichai, you name it. And by the end of that lunch, these executives signed a two-page accord on AI safety. In this accord, the companies agreed to monitor their own AI models, create internal safety teams, bring in independent outside auditors to test their systems, and have an independent board committee overseeing all of it. Now, there's a catch here. None of this is legally binding. This is kind of like a pinky promise that these AI companies are making. President Trump called it morally binding, but I mean, there's no legal penalties if these companies don't follow through. Also, a quick side note, President Trump is also trying to rebrand AI as SI now, and calling it super intelligence instead of artificial intelligence, which I feel like sounds even scarier, but that's just me. Big picture though, I do think there is real incentive for these AI companies to make something like this work, because the alternative could be the government eventually writing the rules for them when it comes to AI safety, which most of these companies don't want. I think a lot of these AI companies would rather self-regulate than have the government do it. And by the way, this is pretty common in other industries. The best example of this is Hollywood. Back in 1968, the studios created their own movie rating system, mainly to keep the government from censoring their films. That's where we got the G, PG, PG-13 and R rating from. That's not something the government mandated. The film industry decided to do that to themselves. And it worked because almost 60 years later, the government still doesn't rate movies. Another somewhat similar example is FINRA, which polices brokers on Wall Street. Now FINRA does have some real teeth. It can find firms and ban brokers, and the SEC is watching over FINRA. So it's not like a perfect example of what these AI companies are doing. Right now, the AI accords look a lot more like the movie rating system than FINRA. You know, it's all voluntary. There's no referee and there's no penalty if these companies break it. The real tension here is that these AI companies are also competing in probably the most intense tech race we've seen in decades. You know, there's a huge demand for the smartest models and the best agents. So if one of these companies slows down in the name of safety, that could give their competitors an advantage. So we'll see if these companies can follow through with this picky promise. Now there was a report in the Wall Street Journal earlier this week that OpenAI delayed the release of their newest model because of safety concerns. The report said that this new model showed higher levels of deception and sometimes used outside tools or services without getting permission from the user first. So OpenAI has delayed the release of that model to fix those problems. By the way, if you want to learn more about AI safety, definitely go check out my interview that I did with Kevin Roose over the weekend. One of his hot takes in that conversation was that he thinks there could be a soft nationalization of one of these AI companies because of a safety incident. I thought that was a wild take, but honestly, I respect it. It was a great conversation with Kevin. I highly recommend you guys checking that out if you want to learn more about AI safety. I mean, Kevin literally wrote a book on it, so he's very much in the loop of what these AI companies are doing. We'll put a link in the description if you guys want to check that out. Sticking with AI, let's talk more about OpenAI because they've had a pretty busy week. Yesterday, the company held their DevDay conference in San Francisco and announced some new features, the big one being an AI agent called DOTS. AI agents are the hot new thing right now, especially with Metamuse having the spotlight for the last few weeks. In fact, Muse is still the number one app on the App Store as of this recording. Well now, OpenAI wants to get in on the action. They launched their own agent called DOTS. It's similar to Muse. It can do things like connect to your calendar and email and help you book flights and research things and write documents, even build software in the background. I got access to DOTS last night. I haven't had much time to play with it, so we'll see if it catches on. My initial impression of it is that it's a lot clunkier to use than Muse is, but I also think that OpenAI is targeting DOTS to be more used by workers and enterprise. In fact, OpenAI also launched a few more tools targeting the enterprise, like their own version of Google Docs. So it's a clear sign that OpenAI is really going after the enterprise market, which I think is smart. I also think that might be bullish for Meta because that might leave them as the default winner for personal AI. Now, the other OpenAI headline from yesterday was a report from Bloomberg that said that OpenAI is trying to raise at least $30 billion at a $1.4 trillion valuation. This new funding round would basically serve as a bridge round until the company IPOs next year. Now, if you've lost track at home, OpenAI just raised $122 billion back in March. So I guess they still need more money to get them through the IPO, which is kind of surprising to me. Anyways, I can't wait for them to IPO because I really want to take a look at what their P&Ls and their margins look like.
4 more minutes of transcript below
Thousands of transcripts fetched by people building searchable podcast archives
Fetch the whole transcript
The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000792409906