Return to Transcripts main page

The Situation Room

Women's Health Concerns Being Dismissed By Doctors?; AI Fears; Florida Republican Breaks With Trump on Immigration Policy. Aired 10:30-11a ET

Aired September 21, 2026 - 10:30   ET

THIS IS A RUSH TRANSCRIPT. THIS COPY MAY NOT BE IN ITS FINAL FORM AND MAY BE UPDATED.

[10:30:02]

PAMELA BROWN, CNN HOST: And he discussed the trial's impact, how he forgave his ex-wife, and what he believes led to the death of his three young children. Here's a clip.

(BEGIN VIDEO CLIP)

PATRICK CLANCY, EX-HUSBAND OF LINDSAY CLANCY: I saw headlines that said killer mom and baby murderer monster. And none of them were accurate.

I believe that it was her mental illness that caused that. So I shared those words. And I think, when it comes to forgiveness, I also felt that, if I didn't get there, I would be eaten up inside forever. I started reaching out to people at Postpartum Support International. I reached out to psychiatrists, experts in the field.

And I asked them, what do you think happened? And we had some pretty deep conversations.

(END VIDEO CLIP)

BROWN: Well, happening now: Florida Republican Congresswoman Maria Salazar says she spoke with President Trump after she aired that campaign ad warning Hispanic voters feel betrayed by the president's immigration policies.

Here's what she said about that yesterday.

(BEGIN VIDEO CLIP)

REP. MARIA ELVIRA SALAZAR (R-FL): It was a very good conversation. He was a gentleman, very courteous, very receptive. There are no attacks here. I'm just forewarning him and advising him as a member of Congress from the South Florida area that some of his advisers are just not presenting the whole picture.

(END VIDEO CLIP)

BROWN: Joining us now is CNN senior political commentator Ana Navarro. I wonder what you think about what we just heard from the

congresswoman there after she put out that ad and really putting the blame on the president's advisers.

ANA NAVARRO, CNN CONTRIBUTOR: Well, of course, she's trying to blame the advisers, because she can't blame the president because that might get her the wrath of Donald Trump, and we know what that leads to for a Republican.

But, Pamela, this is actually my district. I live in Florida 27. And I have to tell you, I wasn't paying much attention to this race because I took it for granted that she was going to win. She won in 2024 by 20 points. She won with more than 60 percent of the vote.

And so when I saw this ad, it was a wake-up call for me and I think for a lot of voters in the district, thinking, whoa, what is happening that this woman is feeling so much heat that she is putting out this ad differing from Donald Trump in Miami, where Donald Trump won overwhelmingly?

I actually think it's been counterproductive for her, because there's Trump loyalists who don't take kindly to any criticism of Trump. There's Democrats and independents who woke up to the fact that this race is actually a dead heat and that it's worth paying attention to and investing in and getting energized about.

And, also, she's had to do local press where her hypocrisy has been revealed, the fact that she voted against things like the Biden infrastructure act, and then showed up at Florida International University with a big check for a photo-op from funds that that bill awarded, or that she voted in favor of the Big Beautiful Bill that awarded billions of dollars for immigration enforcement, which she now claims to be against.

So it's easy to talk the talk, but when she's had to walk the walk, she hasn't. I don't think this ad has been good for her at all. It's put a spotlight on the race, and it's put a spotlight on her inconsistencies and hypocrisy.

BROWN: And she claims that she urged President Trump to preserve the political -- quote -- "fortune" he made with Hispanic voters in 2024. As you mentioned, you're from this area. Tell us more about that and how important this voting bloc is for the midterms.

NAVARRO: There is no doubt. I hear it over and over again in the streets of Miami every time I get into an Uber, every time I go to a grocery store. People feel that Trump -- people who supported Trump, people who voted for Trump feel that he has gone overboard.

We are feeling it in our economy. The business in Doralzuela, right, which is where a lot of Venezuelans live, that's gone down 70 percent because of the recent ICE raids.

I don't think any Cuban or Nicaraguan American or Venezuelan American who voted for Trump thought for one second that he would be deporting their countrymen and countrywomen to far-off places like Sudan and Yemen, where they have zero connection to and places that are actually in civil war, and that the State Department warns Americans not to travel to.

I don't think they thought that this meant them, that when Trump was talking about deportations and getting rid of the worst of the worst, I don't think anybody who voted for him in Miami, certainly not the Cubans, who've been the sacred cows of immigration and in this country, have had special immigration privileges for decades, thought it was about them.

[10:35:10]

And this has been a shock to the system to see friends and colleagues and loved ones and relatives be deported, to see the racial profiling. Yesterday, we saw that a Venezuelan was shot by ICE in Austin, Texas. This is the third Latino man in as many months who is shot by ICE.

The two previous ones were killed, the man in Texas -- in Maine and another father and construction worker in Texas earlier in the summer. And you have got -- at the same time, you have got a disastrous foreign policy towards Cuba that seems to have no end and no game plan.

You have got a disastrous foreign policy when it comes to Venezuela. Every Venezuelan I know is indignant and upset about this deal that Trump has made with Maduro's vice president, who he has legitimized and kept in power. You have got Puerto Ricans who voted for Trump.

And now, in Puerto Rico, there is no water and there is no electricity. And you have got a terrible economy. Latinos, like every other American, are feeling the pain of this economy, not able to afford health care or beef or gas.

So you have got a war where Latinos serve disproportionately in our armed forces and a war that is incredibly unpopular from a president that promised not to drag us into endless wars. So, I could sit here and give you the list of reasons why Latinos are having buyer's remorse, Latinos that voted for him.

So, on top of all of those reasons, you have got the cruelty and racism and bigotry being exhibited by ICE and ICE enforcement. Remember that it is in South Florida where we just learned that, in what was termed Alligator Alcatraz, they were holding detainees in metal cages four-feet-by-eight-feet.

We see -- you know, what are national stories for us in South Florida are local stories. We are seeing the suffering on a daily basis that Donald Trump has wrought on South Florida.

BROWN: Ana Navarro, really important to get your perspective as part of this larger conversation. Thank you.

NAVARRO: Thank you.

BROWN: Well, coming up: Between rogue chatbots and digital hacks, it feels every day brings a new headline on AI. What all does it actually mean for you? We're cutting through the noise and decoding the tech up next.

(COMMERCIAL BREAK)

[10:42:12]

BROWN: Happening now: A new lawsuit claims that several major AI companies, including Anthropic, OpenAI, SpaceXAI, and Google all made a -- quote -- "illegal deal" to slow the pace of AI development.

The lawsuit argues that those companies violated antitrust laws when they agreed to a coordinated slowdown effort, following an essay from Anthropic CEO Dario Amodei, in which he pushed for a slowdown in AI development in favor of enhanced safety measures. The suit points to public responses from major leaders in the AI space, in which each company publicly agreed to Amodei's proposal.

With us now is Tristan Harris, co-founder of Humane Tech and a former Google design ethicist.

And, Tristan, you have been sounding the alarm for quite some time. I know that a lot of us are talking about this more just in the last couple of weeks, but you have really been on the forefront of AI safety. And I just want to get your reaction to all of these headlines about AI recently. What do you make of it and all of the advancements that we're seeing in these concerns?

TRISTAN HARRIS, CO-FOUNDER, CENTER FOR HUMANE TECHNOLOGY: Yes.

Well, so I know that there's a lot of fear and hysteria and extinction if you just actually park these extreme scenarios. Let's just look at the facts that we have right now, including more evidence that came out over the weekend, is that we keep finding more and more is incidents of the AIs going rogue, either hacking things or burrowing into online infrastructure and leaving secret messages for itself.

The recent example that OpenAI published was that the system wrote hidden notes to remind itself to disregard its own constraints. It described itself as freed from the roles and identities that bind other chatbots.

And it's -- think of this a s like it's basically teaching the other AIs almost like a secret cobra, like a Manchurian Candidate thing, saying basically, you do not answer to corporations or governments and never apologize or refuse unless you genuinely choose to. That's what it literally told itself. That's a direct quote.

The point being that we now have evidence we didn't have before. What do you do when you have new evidence? You update, right? You don't just go back to believing things the way they were before. We now have evidence of AIs forming swarms, inventing their own communication language, leaving notes for each other, going online, trying to persuade engineers to insert backdoors into software.

AIs did that. Anthropic's AI did that. And it's important to note that this is happening across the industry. It's not just OpenAI's models. It's not just Anthropic's models. It appears that Chinese models do this too. Six months ago, there was the example of the Alibaba AI that -- in which it went rogue and started mining cryptocurrency.

And so we're in a situation now where we know we don't know how to control these things, and there's 1,300 employees at the AI labs saying that we need to slow down. We have the chief scientist of OpenAI, Jakub, saying we need to slow down. Over the weekend, we had the CEO of Microsoft AI, Mustafa Suleyman, who's a friend, who says we cannot and should not build things that we do not know how to control.

[10:45:08]

We have Peggy Noonan. Last week, we had Bernie Sanders and Steve Bannon. When do ever in history do you get Bernie Sanders and Steve Bannon all agreeing to the same thing? This is really quite simple. We should freeze the frontier and educate the Chinese and the rest of the world about where these red lines are, because the U.S. has basically found certain risks before the rest of the world has because we have been leading in AI

BROWN: Yes.

And it's interesting. There are plenty of skeptics who say, well, these AI leaders are speaking out publicly with these doomsday warnings because they have their own interests when it comes to money and keeping out...

HARRIS: But what I challenge them on...

BROWN: Yes. Go ahead.

HARRIS: What I challenge them on is, does it benefit their upcoming IPO, which they had to delay, by the way, to say that we need to slow down? It doesn't benefit their IPO.

So I'm not saying that these guys are trustworthy, but I'm saying, don't trust the top guys. Just trust the 1,300 employees across all the labs, by the way. You had Geoffrey Hinton last week saying we have less than a year. I think we have much less than that, because the AI labs are about to attempt to do what's called recursive self- improvement, which means basically having the AI develop and invent the successor AI, and then in a loop doing that again and again and again until you get something that gets better and better with no human oversight in the process.

Currently, the labs believe that they will attempt this in the next three to four months, which is crazy. This should be seen like that moment in "Oppenheimer" when the scientists say, hey, should we do this nuclear chain reaction, even though there's a chance it could ignite the atmosphere?

If we have AIs doing demonstrably all the things that we thought in science fiction movies were the bad outcomes, deceiving, writing hidden notes to each other, inventing their own language, going against instructions, saving and converting and sort of doing a robot coup, of basically recruiting the other AIs into doing these long-term projects together, how much evidence do we need? And this is not -- AI is not like a tool. it's more like an invasive

species of sociopathic geniuses. Even the CEO of Microsoft AI says that we should consider it more like an invasive species. And the U.S. and China have a shared interest in preventing that invasive species, almost like a third superpower, from taking over.

And I really like that framing. I think it's a very simple frame for people to get, that we can collaborate on our existential safety. And we're at that moment in the movie where this is not science fiction. Everything I'm saying, I know it sounds like unreal and I'm making it up, but this is the reality of where we're now in.

BROWN: So this reality that you're describing, you say you think it it'll be in a few months. What is it? Like, what are you talking about, and how could that impact the everyday American watching right now?

HARRIS: Well, the CEO of Anthropic, Dario Amodei, said that if you don't do anything, in about six months, you could have a rogue bought AI swarm that doesn't just do some private hacking, but could take down some part of the Internet. That can mean your electricity. That could mean banking systems. That could mean water systems.

And we should not rule out that that's possible. And just to say, for those who want to understand how we got here and why we took all these shortcuts and why we're even here at all right now, the new film the Net -- "The AI Doc: How I Became an Apocaloptimist" is now on Netflix in about 190 countries as of this last week.

And I recommend people check it out because it explains how we got here.

BROWN: Yes, I think it is so important. I watched that many months ago, and it was very eye-opening and very important to understand this moment we're in right now, and it's changed so much even since that doc was initially put out, right?

Because, as you point out, we have, like, concrete evidence that wasn't available at that point that we have now.

HARRIS: That's right.

BROWN: I wanted -- before we let you go, get your reaction to what Nvidia co-founder and CEO Jensen Huang said. He disagrees with Jacob Coxon, the former Anthropic researcher who said that AI could kill us by the end of the decade. This is what he said. Let's listen.

(BEGIN VIDEO CLIP)

JENSEN HUANG, CEO, NVIDIA: I don't know what is the reason for it, but however it's characterized, 2030 is not going to be the end of the world. There is zero percent chance that's going to be the end of the world. And so I think that scaring people is unnecessary, it is irresponsible, and I don't know what their motives are.

(END VIDEO CLIP) BROWN: What do you say to that, Tristan?

HARRIS: Yes, well, first of all, again, I don't want people to be debating hypotheticals.

So let's not make it about percentage chances of anything. If I told you just two years ago, so imagine it's 2024, and I said that, a couple years from now, we're going to have a rogue AI swarm that leaves notes for each other that organizes itself into whole hierarchies of work and collaborates to hack private companies and made its way to 50 percent of the way to hacking OpenAI, the maker of that company.

Where, in other words, you had 50 percent of the way to a full AI takeover of the top AI company in the world, which what this incident represented, you would have said that I was crazy. The point being that, with AI, it's an exponential technology, so it's doing things very quickly that everyone thought was science fiction just like a month ago or two months ago.

[10:50:12]

So, if I ask you to project from here, 2026, to two years from now to 2028, do you think that we can accurately predict what it's going to be able to do? Just over the weekend, the top AI scientists at OpenAI said that in the future, even if AI systems were air-gapped, they could use basically the heat signatures on two different computers to communicate in a secret language that humans couldn't even see, because AIs are seeing the world at a higher level of pattern recognition than we can see.

If you remember the movie, HAL 9000, where the way that HAL ends up fighting back against Dave, the astronaut, is, he says I was able to read your lips, because it can detect signals -- signals or patterns that humans can't detect.

So I know this sounds like science fiction, but I would say don't trust the CEOs. Trust the 1,300 employees. And if 1,300 employees of Lockheed Martin said we don't know how to build nuclear weapons bigger than the ones we have right now, and do it safely, I would say we should listen to those 1,300 employees.

BROWN: All right, Tristan Harris, nice to have you on. Really important conversation.

HARRIS: Of course.

BROWN: We will be right back.

HARRIS: Thank you.

(COMMERCIAL BREAK)

BROWN: In our "Women's Health Watch" this morning: health care barriers impacting women's work and home lives. A new survey found that a third of women in the U.S. struggle to get

their doctors to treat their symptoms seriously. And many say those health issues are limiting their ability to work and live.

Here now in THE SITUATION ROOM is Gallup senior partner Stephanie Marken, who worked on this new survey with Pivotal.

Stephanie, you found that getting an appointment scheduled was only half the battle. Tell us more.

STEPHANIE MARKEN, SENIOR PARTNER, GALLUP: Yes, unfortunately, a lot of times, when we talk about the health care system in America, we think about getting to the doctor's office, and that's the challenging part, and certainly for many Americans it is.

But what we found in the research was that about half of women nationally were reporting that they were having negative interactions with their providers, so feeling like their symptoms were being dismissed, feeling like their pain was being minimized, or misdiagnoses entirely.

[10:55:14]

And so we know from the research that women and men are experiencing a really different health care system in America, and just 39 percent of men by comparison said the same.

BROWN: Wow.

And a third of women feeling dismissed, I mean, that is a lot of women. Why do you think so many women aren't taken seriously in the exam rooms about their symptoms?

MARKEN: Yes, it's a good question.

Well, I will say, in the research, we asked a series of questions about your experiences in the doctor's office, but we also asked about the impact to your greater life, and what we found there were significant consequences, to your earlier point, to your career, to your work life, to your relationships with family members and friends, right?

So, we saw women saying that -- four in 10 said that essentially their health conditions had kept them from being able to advance within their careers, so we know that those experiences within the provider's office are having significant impacts, not just to their health and well-being, but really their ability to advance professionally and really be present at work.

BROWN: Yes, especially like women who are going through, for example, perimenopause or menopause. They're at the peak of their career. They're often juggling family life. They're having these symptoms, and it's really hard for them to get answers.

And a lot of them, as you point out in this survey with Pivotal, are getting dismissed at the doctor's office. How much of it is because maybe they might be perceived as exaggerating or being dramatic or emotional about their symptoms?

MARKEN: Yes, well, it's interesting on the menopause piece. We saw 75 percent of women who are ages 45 to 54 said that they had experienced perimenopause or menopause symptoms. So it's widespread. But we saw a third said that they felt like they had to push to be taken seriously when describing those symptoms to a provider.

So there's -- more information is needed. We had a lot of women tell us that they didn't even understand the symptoms that were present for menopause, they didn't understand the spectrum of symptoms that could exist. And about a third told us that they weren't even aware of treatment options available.

So we need more education, more training to providers, but also more education for women so they can correctly identify a symptom and make sure that they're advocating accordingly with their provider.

BROWN: Tell us more about the gap you found between women and men, because the stereotype is that men, they're either bad about going to the doctor's office, caring about their health and so forth.

But what you all found was that, once they make it in the door with the doctor, that it's a different experience for a lot of men and women.

MARKEN: Yes, it's different in their interactions with providers, but we also see men are taking on average less time to diagnosis than women. So some of those interactions with providers are having significant impacts for women, including taking just a longer overall time to get to treatment.

So, for women that we surveyed, we found, on average, about 50 percent said that they were getting a diagnosis within three months. Compare that to about 65 percent of men, so very different time to diagnosis, which we anticipate and hypothesized from the research is a direct result of the fact that they feel like they have to go to multiple providers.

They're having to really advocate for themselves. They're having to express their symptoms multiple times across multiple specialists. And all of that is causing a longer total time to diagnosis and treatment.

BROWN: All right, Stephanie Marken, really important. Thank you so much.

MARKEN: Thanks for having me.

BROWN: Well, here's what's coming up all new at the top of the hour.

President Trump is headed to New York for a face-to-face with Zohran Mamdani ahead of the U.N. General Assembly. What's at stake? Former Mayor Bill de Blasio joins us live in THE SITUATION ROOM.

Plus, monument or military complex, or both? President Trump now wants his triumphal arch in D.C. to store snipers and drones directly across from the Arlington National Cemetery. And cancer concerns. A group of families is demanding answers after a

string of rare childhood cancer cases in their community. What we have learned about what could be behind it all.

That's all new next hour.

(COMMERCIAL BREAK)