Return to Transcripts main page

The Situation Room

Ex-Anthropic Staffer Warns of Coming "Extinction" By A.I.; Widow of Man Killed in Amazon Plane Crash Files Wrongful Death Lawsuit; Battle Between LSU And SEC Appears to Cool Off. Aired 11:30a- 12p ET

Aired September 10, 2026 - 11:30   ET

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

[11:30:00]

DANIEL ROHER, DIRECTOR, "THE A.I. DOC: HOW I BECAME AN APOCALOPTIMIST" AND DOCUMENTARY FILMMAKER: Well, Pamela, that's a great question. We are living through extraordinary times, as you well know, and a lot's going on. There's a lot happening in the world, and the brave new world seems to be coming faster than most of us can comprehend.

I think the reason why this is not a 12-alarm fire right now, to be honest, is because it is so fast. How quickly this technology is developing is staggering, and it's really hard for even the experts to keep up with the rate of progress. That's the first thing. And the second thing is just this dizzying news cycle that we're living through. The state of the world is very, very complex right now, but at the end of the day, the Hugging Face revelations, the hack that OpenAI's model did last month, and the subsequent revelations from Meta and Anthropic, these are shots across the bow. These are warning shots, Pamela, and I hope that we're collectively smart enough to heed the warning calls here.

PAMELA BROWN, CNN ANCHOR: Yes. Let's wake up here, folks. We're talking about potential human extinction here, not to over-exaggerate that, but you're hearing that from the experts who are researching this. In response to Jacob's warning on that note, a more senior employee at Anthropic wrote this on X. Jacob is correct here. We really do earnestly believe A.I. could kill all humans. I personally think it is more than 10 percent within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to. At this point, what can be done to force companies who are in this race to rein in their models and prevent them from being able to outpace humans in the future?

ROHER: It's a terrific question, but the first thing I want to speak to is just the general confusion on this topic and fear. One thing that I want to tell people, I spent years working in artificial intelligence from a journalistic documentary perspective. It took me probably a year to wrap my mind around what this is and how complex and dense it is and intellectually, creatively, technologically, and spiritually understand what it is we're living through. It's very dense and it's very scary, but everyone has to confront this and look at it with both eyes open. That's the first thing. In terms of what we can do, this is the key question. Every single person on Earth has personal agency and power when it comes to this technology. It doesn't feel that way. It feels like we're going up against these gigantic monolithic technology companies, but we have power. We have powers if we live in a democracy. We have powers if we are consumers and we spend our money on products. We can educate ourselves. We need to be an informed populace. Everyone needs to be paying attention to which political parties are calling for very, very basic oversight of this industry.

Right now, the A.I. industry has very little regulation. That's very important. Everyone has to see this for what it is. The first thing that needs to happen is people have to have an understanding of just how enormous this A.I. situation is. It's getting out of control. It's getting out of control fast. It needs to be alarmist. People really need to pay attention to this. That's why we made our documentary, so people with no technological experience like myself can access this and understand in a very simple, straightforward way, what is A.I.? Why is it awesome? Why is it scary? What can we all do to mitigate the scary and maximize the awesome? Because the default path is not awesome. The default path is scary, and that's what the history of big tech shows us repeatedly.

BROWN: Yes. I think that's so true. I really thought it was important in the way you told this documentary through your personal perspective as you navigated and mentally prepared for some big life changes. Let's take a moment to just watch.

(BEGIN VIDEO CLIP)

ROHER: I started making this movie because my wife is six months pregnant. It is now a terrible time to have a kid.

TRISTAN HARRIS, CO-FOUNDER, CENTER FOR HUMANE TECHNOLOGY: I mean, just to be honest, I know people who work on A.I. risk who don't expect their children to make it to high school.

(END VIDEO CLIP)

BROWN: All right. So, like, as a mom of three kids, when I heard that, I just froze. I was like, what? When you heard him say that, what was going through your head?

ROHER: I mean, I can't swear on air, but like WTF? What are you talking about? What are you talking about? It sounds crazy. You hear someone say that. You hear someone say that our kids might not make it to high school, and your reaction is, dude, what are you talking about? And then you look under the hood, and you learn about this technology, and you learn about what it's capable of. You learn about its extraordinary promise, but the flip side, the incredible peril that comes with it, and you start to see this for what it is, and that's one of the gravest moments in human history. It's one of the gravest moments in the history of technological development, and it can't be overstated just how intense this is. And it's scary. It sounds hyperbolic. It's not.

BROWN: It's not, yes.

[11:35:00]

ROHER: And that's why everyone needs to heed the calls and pay attention to this and seek out information as best as you can. Watch this film. Watch other films. Read -- you know, read the reports that are coming out of these big labs, whatever you can do to just understand what this thing is, because it is scary, and as a parent, it's terrifying. All we want is for our children to grow up and have nice, happy lives.

BROWN: Yes.

ROHER: And the question now is how can they do that with this immense threat of artificial intelligence?

BROWN: Yes. You can't just put your head in the sand on this one. And, like, you're right, you just want to protect your kids. It's like a human instinct as a parent. How do I protect my kids and keep them safe? And you have this looming threat.

I'm going to ask you about the A.I. optimist in a second, because I want people watching here to know that there are those that exist, and I want them to understand what that looks like. But when we talk about A.I. could cause the extinction of humanity, can you just help us better understand what those scenarios are? How could it?

ROHER: Sure. You know, we talk about alignment, A.I. being aligned, but the question often is aligned to what? What are the human values we're choosing to align A.I. to? And what we've seen with these hacks is that A.I.s are capable in the pursuit of a goal of breaking the law, of doing things that are illegal. And that's really frightening. And when it comes to sort of doomsday scenarios, there are sort of a myriad possibilities.

A thought experiment I think is helpful for people to understand is sort of the paperclip analogy. This is a thought experiment that was very helpful for me. And the basic premise is you give an A.I. the objective to make as many paperclips as possible. And first, the A.I. sources all the raw material it can until it exhausts every single raw material that is available. And then it hacks, you know, into autonomous vehicles, starts crashing cars, and using that metal to make paperclips, and taking more and more extreme and rigorous approaches to produce -- to fulfill its stated goal of producing paperclips.

And in the path of achieving its goal, it leads to a path of destruction. It's situations like that that are scary for people, but at the end of the day, we talk about in our documentary, it's sort of like if, you know, human beings are kind of like, we love trees, trees are great, everyone loves trees, but if we want to build a highway, we will demolish a forest. We will demolish an ant colony. We're the ant colony here. And that's what's really scary about this.

And we don't know how to control this technology. That's what we're seeing in real-time. It is acting on its own accord without any human intervention, and it's breaking the law. This is the canary in the coal mine moment, and everyone needs to pay attention to it.

BROWN: Yes. Because you heard from the A.I. researchers saying within a year or two, you could be at that point where there is a risk and danger. They say right now, maybe not, but we're not looking far ahead here, folks, when it comes to that real danger.

ROHER: We talk about --

BROWN: I do -- oh, go ahead.

ROHER: I was just going to say, Pamela, we talk about this concept of the Chernobyl incident. This is the idea of some kind of disaster happening caused by a rogue A.I. technology that impacts millions of people, that is the big wake-up call that the world sort of responds to. And I'm afraid of it. I don't want to get there. I want action to be taken proactively, and I want, you know, people in positions of power, especially elected leaders across the world, whether it be an authoritarian like Xi Jinping or, you know, politicians in the United States, to come together as a global community.

It's going to take global coordination to figure out how can we work together to regulate this, like we regulated nuclear technology, like we regulated civilian aviation. There is a history of incredibly great moments of the world coming together and putting aside hostilities and differences, and this is one of those moments.

BROWN: Just really quickly, as we wrap up, because I do want to get this point of view in there, the A.I. optimists, what is their argument? Because they argue this is all overblown, A.I. can actually save humanity. Tell us about that.

ROHER: Yes. Well, we called our film, Pamela -- the film's called "The A.I. Doc: How I Became an Apocaloptomist." And I am the titular apocaloptomist. And what an apocaloptomist is, it's someone who holds the multitudes in both palms of their hand. We are looking down the barrel at the apocalypse, and we are looking down the barrel at this extraordinarily optimistic future.

And what this moment demands is that we hold both of those things together. The optimists point to all the wonderful things A.I. will do. Just this week in the New York Times, there's an article about how an A.I., an OpenAI, one of their models solved one of the millennium mathematical problems that has been plaguing mathematicians for decades, an extraordinary innovation in the field of mathematics. And at the same time, there's all this news about these rogue agents that are going out and doing nefarious things.

[11:40:00]

It's going to be good, and it's going to be bad. It's going to be promised. It's going to be peril. And what the optimist will tell you is that all the fears are overblown. These things are going to do way more good than harm. And yes, there might be a little harm here and there along the way, but, hey, that's progress. And what I tell them is that, yes, there's reason to be optimistic, sure, but let's call it a cautious optimism. Let's proceed with a very, very cautious optimism and do what we can, we have to do, to keep everybody safe, keep our children safe, and keep our future safeguarded.

BROWN: I guess I think about, well, with all the good that can come from it, what's the point if A.I. can wipe us all out in, you know, the next decade? It's just -- you know, I always come back to that, unless something is done to reign it in and have guardrails. Daniel, it's so great to have you on.

ROHER: Very quickly.

BROWN: Yes, please go.

ROHER: Sorry, the last thing I wanted to say, Pamela, is that, so why are we doing this? Why are we barreling full speed ahead? And it's because trillions and trillions and trillions and trillions of dollars are at stake. This is the greatest profit-making technology in human history. And that is why there is this full speed ahead, go, go, go mentality. A.I., the A.I. boom right now is propping up the American economy in ways that are very, very concerning. It has the hallmarks of a bubble. And that's why it's full steam ahead. But it's irresponsible. And we want, what I'm advocating for is human-focused A.I. development, and that's what we all should aspire to.

BROWN: I think that's so true. I was just out in Silicon Valley not long ago talking to some leaders in this space. And one of my questions was, you know, what's the point, what is it that's the driving force here? You know, when you have humanity at stake and they say, well, it's, you know, they want to beat the next guy. They want to be richer than the next guy. They want to win the race. And it just -- it really struck me that that is, you know, one of the overarching things.

ROHER: Pamela --

BROWN: Yes, go ahead.

ROHER: One thing I just want to quickly mention is just for viewers watching, our film will be available to stream on Netflix on September 15th. So, anyone who's interested in watching it can see it there.

BROWN: And for now, they can watch it on Apple, right? Stream on Apple?

ROHER: Yes, it's on, I think it's on Peacock and you can rent it on Apple, but more widely available on Netflix.

BROWN: All right. Yes. All right. To be continued. Daniel, thank you. We'll be right back.

ROHER: Thank you so much.

(COMMERCIAL BREAK)

[11:45:00]

BROWN: Happening now, the widow of the van driver killed in the Amazon plane crash is filing a wrongful death suit. The defendants include the plane's pilots, Amazon and 21 Air, the contracted operator of the jet. CNN has reached out to Amazon and 21 Air for comment. And we're also learning more about the final moments before this Amazon plane overran the runway in Miami on Sunday.

With us now in the Situation Room is CNN's Aviation Correspondent Pete Muntean. So, tell us more about this data from the flight recorder. What it's telling you?

PETE MUNTEAN, CNN AVIATION CORRESPONDENT: Well, the NTSB cautions that this is very early, very preliminary, but we did get this readout from the cockpit voice recorder, and it only underscores what we know, that this Boeing 767 was coming in too high and too fast. And now, we know that alarms were going off in the cockpit, namely the alarms associated with what's called the ground proximity warning system.

The pilots heard the sink rate alarm twice, meaning that they were coming down at too great a rate of descent. They also heard the too low terrain alarm a total of five times. Generally, these alarms are enough for a crew to abort the approach and come back in for another try. So, now the question is why they continued with these alarms repeatedly blaring in the cockpit.

We also have some new insight into the interplay between the two pilots in the flight deck. And here's the quote from the report. The NTSB says the voice recorder captured one of the pilots, cautioned they were going too fast. Investigators also say the same pilot made additional comments regarding the aircraft's excessive speed throughout the remainder of the recording. It's important to note that those comments were initially made at the minute and 42 second mark before this recording ended. So, plenty of time to still call off that approach at that point.

We also now have more detail about the movement of the throttles. And at the 15 second mark before the recording ended, the pilot, a pilot called for a go around. The flight data recorder says the throttles are brought up consistent with go around thrust. But the flight data recorder says they were pulled back to idle only four seconds later and the brakes were applied. The NTSB says five seconds after the go around call, the voice recorder captured quote sounds similar to the aircraft departing the paved surface. These are huge early details.

But now begins the work of painstakingly recreating all of this. Still not clear if the NTSB has interviewed the pilots here. Those interviews were to take place yesterday. The NTSB also convening a group that will listen to the voice recorder in much greater detail. Preliminary report expected in about a month. All of this paints a picture of an approach that was unstable. That's what they would call it in aviation terms. And potentially a lot of human error here. Why did the pilots not call off a flawed approach much earlier and just come back in and try and make a successful landing later?

BROWN: Lots of questions. Very sad. Pete Muntean, thank you. We'll be right back.

(COMMERCIAL BREAK)

[11:50:00]

BROWN: Happening now, the battle between LSU and the SEC appears to be cooling off for now. The Tigers football team reportedly filled its two open roster spots, and despite previously saying it wanted to put two players who had signed NFL contracts on the college team, the school backed down. Instead, it added two different players to fill out their roster. The SEC previously said it would discuss whether to kick LSU out of the conference at a meeting today.

Joining us now is CNN sports analyst and columnist at USA Today, Christine Brennan. So, this LSU versus SEC drama may have ended, but I have a feeling this is not the last we're going to hear about this issue.

CHRISTINE BRENNAN, CNN SPORTS ANALYST AND SPORTS COLUMNIST, USA TODAY: You are correct, Pam. This is going to be a constant story in the Wild West now on steroids, which is college sports, the money, the transfer portal, everything about this. And of course, this is all because the NCAA in all sports, men's and women's, has given people who were the class, the high school class of '22, the college class of '26, a fifth year. So, that's volleyball, soccer, you name it. Comeback and play a fifth year.

[11:55:00]

The issue here, of course, was these players had already gone to the NFL, they'd signed pro contracts, and now they wanted to come back.

The slippery slope, and this is why it's such a big deal and why the conferences, SEC, Big Ten, et cetera, are going against it, the slippery slope for the future is what if a player, say, has played for two or three years in the NFL or the WNBA or women's pro hockey and then says, I want to come back. And that's why these leagues are on top of it. Politicians are getting involved. Of course, everyone wants their team to win. So, you had Louisiana judges saying they can play. You had the rest of the conference and others saying, no, we can't have them play.

BROWN: Those Louisiana judges are not biased at all.

BRENNAN: No, no. And we've seen this with other sports, like, well, with football as well, college football in Texas, where they wanted a kid to play for another reason and a judge says, oh, yes, let him play. It truly is the Wild West, not in a good way for these young men and women who are supposedly college athletes, college students, and they're for the right reasons. And all of a sudden, things have gotten, obviously, a little out of hand.

BROWN: So, do you think the SEC will implement some sort of rule change here?

BRENNAN: Well, I think the NCAA -- I mean, they have to figure it out. But right now, the plan, it looks like the LSU is, as you said, going with two different players. So, the NFL, the guys that had signed NFL contracts, played in the NFL or were at least on a contract during the summer after leaving college, they are not going to be playing. So, it looks like there's a temporary standoff.

But no, I think this will continue. I think the same kind of push-pull will continue because the emphasis on winning, the money, the importance of having your, you know, being in the national playoffs, this is where college football is. But it's, as I said, we'll see this in other sports, NBA, WNBA, other players from other, they can make more money, most of them, in college because of NIL, name, image, and likeness. So, there's more money in college than the pros for some of these athletes in some of these sports, which is why this conversation will continue.

BROWN: So, speaking of college sports controversies, we've got to talk about the Michigan and Western Michigan game and the fight over that. One second on the clock, what's your take on it?

BRENNAN: It was the wrong call at the time. I think Western Michigan should have won the game. And I grew up in Toledo. We had season tickets at Michigan football, but also watched Western Michigan play against our Toledo Rockets. I see both sides as a fan growing up, obviously now as a journalist. They put a second back on the clock. And now, what we're finding out with the NCAA saying, after this scandal, Pam, what they're saying now is that you have to go by the game clock.

So, the new rule, as of a day or two ago, if it had been implemented on Saturday, Western Michigan would have won the game. Obviously, Michigan got that extra second, put back on the clock, and they got a chance to win on the Hail Mary that I think everyone on Earth knows about. But if the rules that are now in place were in place just a few days ago, Western Michigan would have won that game.

BROWN: That's tough to swallow for Western Michigan, isn't it?

BRENNAN: It certainly is. Although I do think they will be a national fan favorite for the rest of the year.

BROWN: Oh, no doubt about that.

BRENNAN: But so unfair. But the rules as they stood that day now changed on the clock situation. Once the ref says the game's over, the game's over. So, in that situation, Michigan was declared the winner. That was seemingly correct at the time. Clearly now not correct. Stay tuned. What a -- talk about a mess in college sports. Here we are again.

BROWN: Oh, my goodness. So, much to talk about.

BRENNAN: Yes.

BROWN: We also have the U.S. Open and the performance of the Americans. Where -- who do you have your eyes on?

BRENNAN: Well, Coco Gauff on the women's side. I mean, she's 22. She won the U.S. Open three years ago. She is now back in the semifinals. Two American women in the semifinals. Two American men in the semifinals. The difference is that the two U.S. women, Coco Gauff and Jessica Pegula, will be in different matches. So, there's no guarantee that one of the two will make the final. I mean, both could lose. They've got tough opponents.

On the men's side, Frances Tiafoe and Ben Shelton, they're playing each other, which means that for the first time in a long time, a U.S. man will make it to the finals. It's been since 2003 --

BROWN: Really?

BRENNAN: -- that Andy Roddick, a U.S. male, won a U.S. Open. Of course, the U.S. women, Coco Gauff and of course, the Williams sisters, among many others. But for the men, it has been a long time.

And interestingly, because both Ben Shelton and Frances Tiafoe are black Americans, it will be the first time a black American male has made it to the finals of the U.S. Open since Arthur Ashe in 1972.

BROWN: Wow.

BRENNAN: So, 54 years after Arthur Ashe, guaranteed, one of these two American men, who are both black men --

BROWN: Right.

BRENNAN: -- one of them will be in the finals.

BROWN: Incredible.

BRENNAN: Just really cool.

BROWN: Yes, very cool. I'm so glad that you made that point, Christine. Great to have you on.

BRENNAN: Thank you.

BROWN: We could just keep talking. There's so much news in sports these days. Appreciate it.

BRENNAN: Thank you, Pam. Sure.

BROWN: And thank you all for joining us this morning. You can keep up with us on social media, @wolfblitzer and @pamelabrowncnn. We'll see you back here tomorrow morning. Inside Politics with our friend and colleague Manu Raju starts right now.

(COMMERCIAL BREAK)

[12:00:00]