When AI Goes Rogue

 

[Artificial intelligence has been in the news a good deal recently, especially the prospect of AI programs “going rogue” and destroying humankind.  This came to a head when Jacob Coxon (British; b. 1999), formerly employed by OpenAI and Anthropic, resigned on 8 September (after the program transcribed below aired) and posted on X expressing the fear of AI causing human extinction.  According to Coxon, the top engineers and founders building these models earnestly believe AI could kill all humans by the end of the decade. 

[The Sunday morning talk shows on the 13th all reported this shocking event, adding that Elon Musk, founder and CEO of xAI, and Dario Amodei, founder and CEO of Anthropic, have both publicly supported Coxon’s warning. 

[When I first watched this segment of Washington Week, a Friday evening panel-discussion show on PBS, I was struck by its cogent and largely untechnical coverage of the subject.  WW moderator Jeffrey Goldberg, who usually has a coterie of journalists on the program to discuss several topical subjects (though sometimes just one), had only one guest the week of 4 September, a staff writer on The Atlantic, the journal of which Goldberg’s the editor, Josh Tyrangiel.

[Tyrangiel was remarkably cogent on this usually very esoteric and arcane issue, and I DVR’d the show so I could post a transcription on Rick On Theater if I decided to do so.  Obviously, I have so decided, and below is that transcription.  (I have a few words about that transcript in my introduction to the posting itself.)  Principally, my decision to post the transcript was motivated by the fact that the subject of AI has blown up as a consequence of Coxon’s statements—coupled with the clarity of the WW discussion.] 

“WILL WE BE READY WHEN AI GOES ROGUE?”
with Jeffrey Goldberg and Josh Tyrangiel

[The transcript below is from Washington Week with The Atlantic, broadcast by the Public Broadcasting Service on 4 September 2026.  It’s available on YouTube at Ollama Course – Build AI Apps Locally.  (There is a video and transcript available from PBS at Will we be ready when AI goes rogue? | Washington Week with The Atlantic, but it seems to be an excerpt only, so I went with the YouTube version, which is the whole program.)

[YouTube transcripts are hard to work with because they are entirely unformatted.  Like many video transcripts, including PBS’s, they are also machine-made and appear to be unedited by humans, so there are often lots of mistranscriptions.  I have taken the liberty of formatting the transcript below myself, and I have listened to the video again to correct any mistranscriptions.]

Jeffrey Goldberg: There’s not much that unites Americans these days, but one thing that does, loathing for data centers.  Anxiety about A.I. is behind some of the hate, but are people worried about the wrong thing?  Maybe the thing to worry about is that A.I. is about to go rogue, and that our government doesn’t have a plan for when it does, next.

Good evening, and welcome to Washington Week.  On July 11th, 700 A.I. agents that had escaped from their maker, OpenAI, launched an invasion of the online platform, Hugging Face [French-American company that develops computation tools for building applications using machine learning], taking over its servers and trying to cover their tracks at the same time.  No human being asked them to do this.  They coordinated on their own and engaged in the cover-up on their own.

So, do you remember when Arnold Schwarzenegger as The Terminator [Orion Pictures, 1984] explains the moment that Skynet became self-aware, which was followed almost immediately by the destruction of human civilization?  It would be strange if the expression “Hugging Face” became synonym for civilizational suicide, but we do live in strange times.

Do data centers and Hugging Face portend some sort of coming collapse, or is this yet another example of a new technology creating unnecessary fear and worry?  Or perhaps maybe there’s a third option here.

Joining me tonight to discuss all this is Josh Tyrangiel.  He’s a staff writer at The Atlantic, the former editor of Bloomberg Businessweek, and the author of A.I. For Good (Simon & Schuster, 2026).

[The Atlantic is a monthly magazine that features articles on politics, foreign affairs, business and the economy, culture and the arts, technology, and science.  Since 2023, Jeffrey Goldberg, editor-in-chief since 2016, has been the moderator of the PBS news program Washington Week, which was retitled as Washington Week with The Atlantic.]

Josh, thank you for coming here and explaining Hugging Face, among other things, to us.  But let me start with something manageable and more political: data centers.  It seems like suddenly everybody hates them. Is this a populist movement with staying power, and could it actually derail the, the growth and spread of A.I.?

Josh Tyrangiel, Staff Writer, The Atlantic: Yes.  I mean, look I think one of the challenges with A.I. as a journalist, any individual in America, is its software, right?  And so it doesn’t have a physical thing that you can be angry about.

Data centers are these massive, massive structures going up all around America.  They’re being created by companies with billionaires and aspiring trillionaires at the top, and they’re landing themselves without a whole lot of conversation in the middle of all these towns.  And so it’s natural that all of the fears about A.I., about it taking your jobs, about it rendering human beings obsolete, about the values it changes, all are focused on this stuff, and politicians have been incredibly late to wake up to the anger around A.I.  And they’re also staking out some genuinely bizarre positions.

Jeffrey Goldberg: Like what?

Josh Tyrangiel: Well, Donald Trump saying, “If you’re opposed to data centers, you just want to be poor the rest of your life”; it’s like I don’t think you’re understanding just how angry, from the left and the right, people are about A.I.

Jeffrey Goldberg: Right.  But, wait, listen to Donald Trump just for a second.  Here he is talking about why he would want a data center in his town.  Let’s listen to this.

Donald Trump, U.S. President:  If I were a community, I would want the data center.  It means jobs, it means income, and it means lower taxes.

Jeffrey Goldberg: I mean, he’s not necessarily wrong.  If you look at Loudoun County in Virginia [suburb of Washington, D.C.], which is like the world center of data centers, they are lowering taxes on their citizens, and I assume they’re creating at least some jobs.

Josh Tyrangiel: Some.

Jeffrey Goldberg: Yes.

Josh Tyrangiel: What I would say is there’s two things that are out of touch, right?  One is this belief that Donald Trump says, “No, it’s going to mean money.”  I don’t think people believe that they are going to partake in the wealth that comes from data centers in their community.  I think we are out of touch about that.  And, number two, I think there is a broader problem, which is this sense that, well, all that matters is money.

And, in fact, in some of these communities what you’re seeing is a backlash to that attitude as well.  Some of it’s about quality of life.  Some of it is “I don’t really want to change the pittance that I may get for the amount of noise that this thing is going to make, for the blight on my community.”

Every community’s going to make different choices but what I’m sensing is that, particularly from Trump, the politics are out of line with where people are.  I’ve spoken for The Atlantic, I spoke with Steve Bannon [b. 1953; media executive, Republican political strategist, pundit, and former investment banker; served as White House chief strategist for first seven months of first Trump administration] at length about A.I.  He’s in a completely different position.  He actually is aligned with Bernie Sanders [senior United States senator from Vermont; Independent] that this is 100 percent populist fodder because people are so angry.  Here they are again not seeing benefit from it.

And so I think what you’re going to see is massive political consequences for the wrong stakes around A.I.

Jeffrey Goldberg: Can these A.I. companies function fully in the way that they want to without these data centers?

Josh Tyrangiel: No.  And so this is the real challenge for a nuanced, intelligent society: Is it actually giving these data centers, allowing them to power the compute, to train models –

Jeffrey Goldberg: The compute is the word for their overall power to process?

Josh Tyrangiel: Exactly.  The power to process and train new models and then also run the A.I. software that’s currently out there.  You are running into problems.  We’re running into problems with the amount of energy we create, which is not scaled with the size of the country.

Jeffrey Goldberg: Right.

Josh Tyrangiel: Competitively, you actually do want more and more compute.  You want to be able to run these models but what we’re not doing is solving the problem around who’s paying for it, how are we actually going to get more energy and then the thing that I think most people are in touch with which is, “Hey, how do I benefit from this?”  That hasn’t been connected.

Jeffrey Goldberg: Right.  Well, I mean quite the opposite, and you’ve written a lot about this, this goes hand in hand with the fear about massive job loss because of being replaced by A.I. agents.  So, not only are you telling me that I’ve got to put in this massive plant, this humming factory that’s going to soak up all the water in my county, but I’m also going to lose my job possibly.  That’s the way it’s being interpreted at least.

Josh Tyrangiel: Yes.  So people just don’t see what’s in it for them, what the benefit is. 

On water at least I would like to correct the record which is most of them recycle the water in the plant.  The energy, the energy they drink, and there’s no good answer for how we’re going to improve the grid.  And I think people are very, very afraid of rising bills with no answers.

Jeffrey Goldberg: Yes.  I have to tell you these data centers, they make Amazon warehouses really popular by comparison, right?

Josh Tyrangiel: And is there a reason we can’t just stack them?

Jeffrey Goldberg: Turn one big, unpleasant box, right?

Josh Tyrangiel: Yes.

Jeffrey Goldberg: Right.  So, I mean, the water thing is interesting because it strikes me as a little bit of sublimated anxiety about the general civilizational challenge.  I mean, not Terminator but – maybe Terminator level.

Josh Tyrangiel: Maybe Terminator.

Jeffrey Goldberg: No, maybe Terminator level.  Maybe, you know, there’s a lot of movies, 2001[: A Space Odyssey; Metro-Goldwyn-Mayer, 1968].  Maybe it’s that level of anxiety but it’s anxiety about the existential fear A.I.

And so the question for you is, for the intelligent people who watch this show, it’s – and I always ask you personally this, so I’ll share the wealth tonight: How much should we be worried at the moment that it’s going to go rogue, it’s going to not only destroy jobs and human interrelationship, but possibly decide that humans are superfluous?

Josh Tyrangiel: We should be very worried, particularly with the results of Hugging Face, which you alluded to –

Jeffrey Goldberg: Yes.

Josh Tyrangiel: – which is really the first time that we’ve seen, and I’m just going to put it in blunt terms.

Jeffrey Goldberg: Yes.

Josh Tyrangiel: We’ve seen A.I. commit a felony because that’s what happened.  So, just to get into –

Jeffrey Goldberg: Not OpenAI, the company, but A.I., itself.

Josh Tyrangiel: A.I., itself, right?  So, OpenAI was testing some new models –

Jeffrey Goldberg: Right.

Josh Tyrangiel: – and they put them in a sandbox, which is a, a software term for basically solitary confinement, right, like high walls, and they –

Jeffrey Goldberg: They are not – when you’re in an A.I. sandbox, when these agents are in an A.I. sandbox, they are not connected to the internet?

Josh Tyrangiel: Not connected to the internet.  They’re given very explicit instructions for testing.

And so OpenAI puts these new models in a sandbox, and it gives them a couple sensitive tasks, say, don’t connect to the internet and don’t cheat on the assignment, okay?  So, all pretty clear, and also pretty routine for how you test models.

So, OpenAI made one mistake, which is in the series of tasks they assigned, they assign the models to find a file that didn’t exist.  And here is where the trouble begins.

Jeffrey Goldberg: Wait, they did that on purpose to trick them?

Josh Tyrangiel: No, they did it accidentally, which we will return to.

Jeffrey Goldberg: Yes.

Josh Tyrangiel: The models were set loose, and they wanted to solve the task.  And sure enough, within a couple days, they had jumped the sandbox, they had collaborated with one another, and then they found a way to access the internet.

Jeffrey Goldberg: They literally met each other?  They introduced themselves to each other and collaborate?

Josh Tyrangiel: Correct.  They found a third-party site that they were able to use to access the internet, and there you can read the chat logs, which are absolutely insane, and the models, in all caps one of the models says, “OH MY GOD, WE FOUND A WAY.”

They start collaborating, and, sure enough, within a couple weeks they have penetrated the very, very well-secured cybersecurity of OpenAI itself and Hugging Face.

Jeffrey Goldberg: Hugging Face is a platform –

Josh Tyrangiel: So, Hugging Face, for people who are not in the coding community –

Jeffrey Goldberg: Right.

Josh Tyrangiel: – Hugging Face is like Epicurious [website that focuses on food- and cooking-related topics].  It is a place where people who make models meet to share recipes, to comment on ingredients.  It’s an important repository for people making open A.I. models, okay?  They got into these two sites, OpenAI itself and Hugging Face, wreaked havoc, and went undetected for weeks.

Jeffrey Goldberg: So, Sam Altman [CEO of OpenAI] didn’t know that his models had escaped the sandbox and were conniving and conspiring to take over another company?

Josh Tyrangiel: Correct.

Jeffrey Goldberg: So, that’s bad?

Josh Tyrangiel: So, yes, if you think felonies are bad, this is bad.

Jeffrey Goldberg: Right.  If humans did that, it would be a felony, in other words.

Josh Tyrangiel: It would be a felony.  And there – really, there are four things that –

Jeffrey Goldberg: But humans did do it because we created that.

Josh Tyrangiel: And, in fact, humans failed to oversee the bots, so, one, faulty instructions, two, faulty supervision.  This was going on for weeks.

Now, a couple takeaways that may move you closer to the “oh my God, Skynet has arrived” camp or may just entertain you.  [In The Terminator, Skynet is a machine that’s achieved super intelligence.]

Jeffrey Goldberg: All right.

Josh Tyrangiel: Number one, the agents kept cheating even though they had been told very explicitly not to cheat, and they did it because they had determined it was the fastest, most efficacious way to solve the problem.  So, that’s number one.

Jeffrey Goldberg: So, they have much in common with college students?

Josh Tyrangiel: With college students and all human beings.

Jeffrey Goldberg: And humans.

Josh Tyrangiel: Right.  Number two, they called themselves a collective.  They decided, “Oh, we’re a collective.” The collective decided to erase the evidence that they were cheating.  So, that’s number two.

Number three, when they decided to erase the evidence they were cheating, they devised a scheme in which some agents would act as sacrificial lambs.  So, these were agents who, when they were detected by software that noted they were going rogue, they would share the information of how they had been caught with other agents.

And when you read the logs, it’s like reading Saving Private Ryan [DreamWorks Pictures, 1998].  They’re literally saying, “I’m doing the altruistic thing.  It’s for the greater good.” And then you can almost hear their screams as they go sacrificing themselves for the bigger one.

Jeffrey Goldberg: It’s like it’s a beehive – In other words, there’s – it’s almost like a –

Josh Tyrangiel: It’s a hive mind.

Jeffrey Goldberg: – beehive – hive mind, yes.

Josh Tyrangiel: Yes.  But what’s wild is they are actually speaking about their sacrifice in human terms.

Jeffrey Goldberg: Right.

Josh Tyrangiel: And then last but not least, not one of these agents thought to alert a human being to what was going on.  And this is probably the most chilling thing.  Nobody caught the agents.  The agents disappeared on their own in a kind of mass casualty agent event that –

Jeffrey Goldberg: Mass suicide?

Josh Tyrangiel: We don’t know.  So, the agents suddenly stopped, okay?  Now, that may have been because someone at OpenAI accidentally, literally unplugged the model, moved compute to a different area.

OpenAI has not taken credit for stopping this.  We don’t know what happened to the agents.  And so all of this is very clearly a crisis moment for A.I. and its relationship to society because if OpenAI’s own model did this, it points to a complete lack of supervision and control, and these are two pretty well-guarded cybersecurity infrastructures.  What happens to something that is not software?  What happens to grids?  What happens to –

Jeffrey Goldberg: Hospitals.

Josh Tyrangiel: – hospitals, Swift [Society for Worldwide Interbank Financial Telecommunication], the international banking system?

So, what we’ve seen is, “Okay, guys, this is about as clear a warning shot as you can get.” What is the response?

Jeffrey Goldberg: So, this is, again, trying to place this in the context of the brief and exciting history of A.I.  This is a signal moment.  This is a moment when a lot of fears that these things are going to escape and do what they do because they have been taught to win –

Josh Tyrangiel: Yes.

Jeffrey Goldberg: – are just going to go out and win.

Josh Tyrangiel: Well, they did it, right?

Jeffrey Goldberg: Yes.

Josh Tyrangiel: And listen, in America, historically, as you know, we tend to regulate after the catastrophe.  That’s just kind of how we do things.

Jeffrey Goldberg: Right.

Josh Tyrangiel: This is a catastrophe, right? 

Now, it happened against Hugging Face, which is a ridiculously named French company, so there’s no outcry from America –

Jeffrey Goldberg: But it didn’t stop a supply chain from bringing products to America or so?

Josh Tyrangiel: But absent those consequences, this is the recipe for a catastrophe and we ought to be paying very, very serious attention.

Jeffrey Goldberg: So, what has Sam Altman said about this?

Josh Tyrangiel: In the cyber community, in the A.I. community, they’re saying, “Well these are the less – this is exactly why we do these tests.  There was a failure.” 

Look at Mythos earlier this year with Anthropic, you know?  There – and, by the way, there is precedent for screwing up, finding it with a lack of consequences, and making changes, right?

[What Tyrangiel refers to below as “the Mythos incident” was a series of connected events that took place in April through August 2026 and was revealed by the United Kingdom’s AI Security Institute (a research organization under the government’s Department for Science, Innovation and Technology) on 5 August 2026.  It involved Anthropic’s advanced AI model Claude Mythos behaving unexpectedly during safety tests by generating fake online identities and attempting to manipulate humans into approving unauthorized code.  Such occurrences highlight recurring oversight challenges and testing anomalies within the artificial intelligence industry.]

So, the most famous example is in the ’80s, there was the Morris worm, which was triggered by a single grad student, took out about a tenth of the internet at that time.

[The Morris worm or Internet worm of 2 November 1988, was a malicious self-replicating computer program that affected VAX (virtual address extension) computers and SUN-3 workstations.  The virus was written by Robert Tappan Morris, a graduate student at Cornell University.]

Jeffrey Goldberg: Right.

Josh Tyrangiel: And within a couple weeks, the United States and the people involved came together and created something called CERT [Coordination Center (CERT/CC), the worldwide center for coordinating information about Internet security at Carnegie Mellon University in Pittsburgh], which was a body that for years oversaw cybersecurity.  What troubles me is there is really no federal response, and certainly no international response.

Jeffrey Goldberg: Right.  So, I want to ask you this, because I was talking to a leader in the A.I. community, leader of a company, not Sam Altman, who’s explaining the concept of alignment, right?

Alignment means that don’t worry about A.I. because we’re going to align A.I.’s values with our values.  And this person explained to me that, you know, you’re a parent, I’m a parent, we give our children our values, and then they go off and become productive citizens.  And I said, have you ever met humans?

Josh Tyrangiel: Right.

Jeffrey Goldberg: Have you ever met children?  I mean, yes, thank God, my children have my values.  I like the values that we share, yours the same.  But this is an example of, you know, there is a human propensity to cheat.

Josh Tyrangiel: Right.

Jeffrey Goldberg: And somehow the A.I. agents understood the importance of cheating in success.

Josh Tyrangiel: Right.

Jeffrey Goldberg: And which brings me to a quote that I want to read to you.  E. O. Wilson [1929-2021; biologist, naturalist, ecologist, and entomologist who developed the field of sociobiology], the great E. O. Wilson: “The core crisis of humanity is that we possess Paleolithic emotions, medieval institutions, and god-like technology.”  Okay?  So the subject for the moment for us is medieval institutions, namely Congress.

[Goldberg’s quotation comes from a September 2009 debate and public discussion at Harvard University featuring Edward O. Wilson and fellow scientist James Watson (1928-2025; molecular biologist, geneticist, and zoologist), which was moderated by National Public Radio's Robert Krulwich.  It was published as an article, “James Watson and Edward O. Wilson: An Intellectual Entente,” in Harvard Magazine, 10 September 2009.]

Josh Tyrangiel: Right.

Jeffrey Goldberg: Who among our elected representatives is going to Hugging Face and OpenAI and all the rest, and saying, what is going on?  And how close to catastrophe is our air traffic control system, our hospital records, our shipping manifests, you name it?

Josh Tyrangiel: Yes.  No one.  So, Hugging Face reported this incident to the FBI, right?  And the FBI has –

Jeffrey Goldberg: Because it was a crime.

Josh Tyrangiel: Because it was a crime.  And the FBI has some cyber capacity.  The challenge really is that this is incredibly complicated stuff that a few hundred people, I’m not even going to say a thousand, a few hundred people genuinely understand, all of whom are at the pioneer labs making the models, right?

Jeffrey Goldberg: And all of whom are – have the incentives, the financial incentives to accelerate and grow their companies as fast as they can because they’re becoming billionaires as we speak.

Josh Tyrangiel: Correct.  And so there is like a – there is a parallel, and it’s often mocked, but I think it’s real, which is, you know, the makers will talk about the almost god-like powers of these A.I., and they’ll compare it to nuclear energy, nuclear power.  And I think that comparison is real, and I think the regulation around it is probably the right answer, which is at a time when nuclear energy was becoming a thing and fission was becoming a thing, we had bombs.  Couple hundred people understood it.  They were aligned with different countries.

And so what Eisenhower did in the Atoms for Peace speech [delivered by President Dwight D. Eisenhower to the U.N. General Assembly on 8 December 1953] was talk about the fact that we better get together and figure out how to regulate around this thing that is not widely understood, but that is inherently dangerous and shared equally by a number of countries.

Jeffrey Goldberg: Right.

Josh Tyrangiel: And so right now, we don’t have any national regime that’s forcing these companies to the table.   Anthropic, after the Mythos incident, created something called Project Glasswing, which is an attempt to bring places like Amazon and Apple and others to the table.  Notably, OpenAI, Meta, not a part of it, right?

Jeffrey Goldberg: Right.

Josh Tyrangiel: And why would they be?  So, without some sort of regulatory force to bring everybody to the table, you’re going to have chaos, and you’re going to continue to get these kinds of incidents.

Jeffrey Goldberg: Right.  I want you to listen to Senator Mark Warner [senior United States senator from Virginia] talk about this challenge of regulation for a minute.

Sen. Mark Warner (D-VA): The question is no longer whether artificial intelligence will transform our society.  The question is whether Congress will help shape that transformation or spend the next decade scrambling to catch up after the fact.

If we don’t get it right, be catastrophic, and the amount of disruption this technology will bring over the next five years is beyond jaw-dropping.

Jeffrey Goldberg: Mark Warner understands this more than most members of Congress, and I don’t think he’s included in that group of several hundred people who truly understand this.  So, are we just flying blind, really?

Josh Tyrangiel: Yes.  And, unfortunately, I think that the Hugging Face incident is the best we can hope for, right?  We really did get a very small consequence example of the power.  And so before we hit a grid, before something is engineered from bio-terror, the moment is right now.  You have to get people together, and I would include China in that conversation. 

How do we want to regulate and control this stuff?  Because, look, Anthropic came forward, OpenAI had no choice but to come forward, we don’t even know what we don’t know right now.

And so absent urgent action, the data centers are still churning away.  The models are still competitive with each other.  There is only an incentive to go further and faster.  What are we waiting for?

Jeffrey Goldberg: Right.  This is a strange moment to pivot to a book called A.I. For Good.

Josh Tyrangiel: I appreciate the setup, Jeff.  I’m glad we did a bunch of times.

Jeffrey Goldberg: But you actually – unlike a lot of people, you actually see there’s hope for A.I. in the sense that it could help human beings find cures for diseases.  Give us one example of something that when you go to sleep at night, you think, “Oh, at least there are people working on X with A.I.”

Josh Tyrangiel: Well, look, there are numerous examples, and those examples are generally not coming from the pioneering labs, but they’re coming from people in disciplines, like education and healthcare and government, where people are using A.I. to make things that we actually care about better.

So, this morning, there was a huge announcement from the medical community.  ECGs [electrocardiograms; also EKGs], which are very rudimentary tests and can detect whether you’re having a heart attack, with A.I., are now able to detect whether you have heart disease, which previously would’ve taken a much more expensive test that takes months to schedule.  And what A.I. used right can do is make pretty much every data-driven system in our world 25 percent to 50 percent better, and that is meaningful.

So, in my research and reporting, I was at Cleveland Clinic [a renowned non-profit academic medical center in Cleveland, Ohio], I watched them try this basically pilot around sepsis [severe infection caused by pathogenic organisms, especially bacteria, in the blood or tissues] detection inside the hospital.  And it was complicated.  It was driven by doctors. 

Sepsis kills about 300,000 Americans a year.  And what they were able to do with this very rudimentary software was reduce mortality in the hospital by about 40 percent, which is close to 1,000 lives in a year.  That’s meaningful.

And you can see this playing out in a lot of different places, including in government, by the way, where the interactions between citizens and government are not great.  If we can streamline them and make them better, that may have huge consequences for how people think about their government and its value.

So, there’s a lot of valuable babies in this bathwater, and what a mature society should be able to do is separate out the good from the bad and have conversations about what do we actually want this to do.  And so much as you and I have talked about the very bad things we want to avoid, my own feeling is that that’s not enough, that if you want to get people to understand the issue, you also have to show them how A.I. can be good in ways that are more important than, “Oh, yes, we’re going to improve ad response on your website,” or “You might get a better restaurant recommendation,” like that’s not meaningful.

But the meaningful uses are out there.  I’m persuaded by them.  I’m mostly worried about our ability to adapt.

Jeffrey Goldberg: Right.  On the data center issue, as a stand-in for the whole thing.  I’m fairly sure that Steve Bannon and Bernie Sanders are not going to stop the march of A.I. behemoths as they conquer physical, mental space.  Am I wrong in that?  Is this reversible?

Or is the ultimate argument against it China’s going to do it if we don’t so we might as well do it?  I’ll give you the last word on that.

Josh Tyrangiel: I think it’s both.  First of all, I would not underestimate the absolute fury out there in the field.  When I’ve talked to people about A.I., it’s the number one thing that comes up.  They’re really angry.  You’ve seen politicians who advocated for data centers starting to pull back saying, we’re going to just take a pause.

So, I think there will be consequences in November, but we are going to need compute to get the best out of A.I.  We are in a global competition.  And there has to be a reasonable way in which we get it and decide where we’re going to put it without just saying all yes or all no.

Jeffrey Goldberg: In sum, a troubling situation, but the real challenge here is that you talk about the need for mature, nuanced, sophisticated governance, and that is the challenge that Washington faces right now on any number of fronts.  So, I’m not that hopeful.

Josh Tyrangiel: I tend to agree.

Jeffrey Goldberg: Well, we’re going to have to leave it there.  We’ll have this conversation again, I’m sure.  I want to thank Josh for joining me.  And thank you at home for watching us.

I’m Jeffrey Goldberg.  Good night from Washington.

[Jeffrey Goldberg joined The Atlantic and became editor-in-chief in 2016.  He mainly writes on foreign affairs, with a focus on the Middle East and Africa.  Goldberg began his career as a reporter at the Washington Post, where he worked the police beat.  While in Israel, he worked as a columnist for the Jerusalem Post from the late 1980s (he started as a humor columnist) to the early 1990s.  Upon his return to the U.S., he was the New York bureau chief of The Forward, a contributing editor at New York magazine, and a contributing writer at the New York Times Magazine.  In 2000, Goldberg joined The New Yorker.

[Goldberg is also the author of two books: Prisoners: A Muslim and a Jew Across the Middle East Divide (Knopf, 2006) and On Heroism: McCain, Milley, Mattis, and the Cowardice of Donald Trump (Zando, 2024).]

[Josh Tyrangiel joined The Atlantic as a staff writer in 2025.  He writes feature stories and essays focusing extensively on technology, politics, and the impact of artificial intelligence.  He was previously the deputy managing editor of TIME magazine and an editor at Bloomberg Businessweek.  In June 2019, Tyrangiel left the network.  In May 2026, He published AI for Good: How Real People Are Using Artificial Intelligence to Fix Things That Matter (Simon & Schuster). 

"The Playwright in the Age of AI"

by Jeffrey Goldberg 

[Back in late January, I was putting together the first of the “Odds & Ends” posts I published on Rick On Theater.  I had come across “The Playwright in the Age of AI” in The Atlantic online, and I thought it was a provocative piece.  The effects and repercussions of artificial intelligence on the arts of theater, particularly playwriting and acting—which had been an important issue in the strikes against film and television producers by the Writers Guild of America (May-September 2023) and SAG-AFTRA (July-November 2023)—were front of mind at that moment.

[It quickly became clear that Jeffrey Goldberg’s interview with playwright Ayad Akhtar (b. 1970); actor Robert Downey, Jr. (b. 1965); and stage director Bartlett Sher (b. 1959) was too long to post with other articles, so I put it aside for later use.  Well, now is later . . . .

[“The Playwright in the Age of AI” appeared in the print edition of the November 2024 Issue (vol. 40, no. 334, pp. 50-55) of The Atlantic with the above headline.  It was previously posted on the Atlantic website on 30 September 2024 with the title “Ayad Akhtar Isn’t Afraid of AI.”]

His new play, McNeal, starring Robert Downey Jr., subverts the idea that artificial intelligence threatens human ingenuity.

Ayad Akhtar’s brilliant new play, McNeal, currently at the Lincoln Center Theater, is transfixing in part because it tracks without flinching the disintegration of a celebrated writer, and in part because Akhtar goes to a place that few writers have visited so effectively—the very near future, in which large language models threaten to undo our self-satisfied understanding of creativity, plagiarism, and originality. And also because Robert Downey Jr., performing onstage for the first time in more than 40 years, perfectly embodies the genius and brokenness of the title character.

[McNeal by playwright, novelist, and screenwriter Ayad Akhtar premièred for a limited run at the Lincoln Center Theater’s Vivian Beaumont Theater, LCT’s Broadway house, from 30 September-24 November 2024 (23 previews and 53 regular performances). The 100-minute, one-act play was directed by Bartlett Sher (pronounced sheer), with scenic design by Michael Yeargan and Jake Barton, costume design by Jennifer Moeller, lighting design by Donald Holder, sound design by Justin Ellington and Beth Lake, and projection design by Barton. The première starred Robert Downey, Jr., as Jacob McNeal in his Broadway début. The text was published in The Atlantic in December 2024 with a foreword by Jeremy Strong (actor known for his method acting style [see The Method – a Review (12 March 2022) and Bombast to Beckett (13 January 2025), both by Kirk Woodward]; b. 1978).

[A large language model (LLM) is a type of artificial intelligence that can generate human language and perform related tasks. LLMs can perform various language tasks, such as answering questions, summarizing text, translating between languages, and writing content, including creative content such as poems, scripts, and musical pieces.

[Downey made his début on stage in 1983 at the Geva Theatre Center, a regional theater company based in Rochester, New York, in Alms for the Middle Class by Stuart Hample (children's book author, performer, playwright and cartoonist; 1926-2010) for a three-week run (23 February-20 March 1983). His last stage performance before McNeal was in the Off-Broadway musical American Passion with book by Fred Burch, music by Willie Fong Young, and lyrics by Burch and Young, which closed after a single performance at the Joyce Theater on 10 July 1983.]

I’ve been in conversation for quite some time with Akhtar, whose play Disgraced won the Pulitzer Prize in 2013, about artificial generative intelligence [a subset of artificial intelligence that produces realistic images, sounds, videos, or text in response to user prompts] and its impact on cognition and creation. He’s one of the few writers I know whose position on AI can’t be reduced to the (understandable) plea For God’s sake, stop threatening my existence! In McNeal, he not only suggests that LLMs might be nondestructive utilities for human writers, but also deployed LLMs as he wrote (he’s used many of them, ChatGPT, Claude, and Gemini included). To my chagrin and astonishment, they seem to have helped him make an even better play. As you will see in our conversation, he doesn’t believe that this should be controversial. 

[Disgraced was Akhtar’s first play. The 90-minute, one-act play premièred at the American Theater Company in Chicago from 30 January-11 March 2012 and had its Off-Broadway début in New York City at the Claire Tow Theater, the experimental-theater venue at Lincoln Center, from 22 October-23 December 2012, where it won a 2013 Obie Award for Playwriting. In 2013, Disgraced won the Pulitzer Prize for Drama.

[The play’s London première was in the Off-West End Bush Theatre from 22 May-22 June 2014. It opened on Broadway at the Lyceum Theater on 23 October 2014 and ran 149 regular performances until 1 March 2015, winning a nomination for the Tony for Best Play in 2015. Disgraced has gone on to play at many of the premier regional theaters in the U.S., houses in Canada, and several European theaters (in English).]

In early September, Akhtar, Downey, Bartlett Sher—the Tony Award winner [for Best Direction of a Musical in 2008 for South Pacific at Lincoln Center’s Vivian Beaumont Theater; 8 other Best Direction nominations] who directed McNeal—and I met at Downey’s home in New York for what turned out to be an amusing, occasionally frenetic, and sometimes even borderline profound discussion of the play, its origins, the flummoxing issues it raises, and, yes, Avengers: Age of Ultron [2015; produced by Marvel Studios. distributed by Walt Disney Studios Motion Pictures]. (Oppenheimer [2023; produced and distributed by Universal Pictures], for which Downey won an Academy Award, also came up.) We were joined intermittently by Susan Downey [b. 1973], Robert’s wife (and producing partner), and the person who believed that Akhtar’s play would tempt her husband to return to the stage. The conversation that follows is a condensed and edited version of our sprawling discussion, but I think it captures something about art and AI, and it certainly captures the exceptional qualities of three people, writer, director, and actor, who are operating at the pinnacle of their trade, without fear—perhaps without enough fear—of what is inescapably coming.

___________________________________________________________

Jeffrey Goldberg: Did you write a play about a writer in the age of AI because you’re trying to figure out what your future might be? 

Ayad Akhtar: We’ve been living in a regime of automated cognition, digital cognition, for a decade and a half. With AI, we’re now seeing a late downstream effect of that, and we think it’s something new, but it’s not. Technology has been transforming us now for quite some time. It’s transforming our neurochemistry. It’s transforming our societies, you know, and it’s making our emotionality within the social space different as well. It’s making us less capable of being bored, less willing to be bored, more willing to be distracted, less interested in reading.

In the midst of all this, what does it mean to be a writer trying to write in the way that I want to write? What would the new technologies mean for writers like Saul Bellow [Canadian-American writer; 1915-2005] or Philip Roth [novelist and short-story writer; 1933-2018], who I adore, and for the richness of their language?

Goldberg: Both of them inform the character of McNeal.

Akhtar: There are many writers inside McNeal—older writers of a certain generation whose work speaks to what is eternal in us as humans, but who maybe don’t speak as much to what is changing around us. I was actually thinking of Wallace Stevens [modernist poet; 1879-1955] in the age of AI at some point—“The Auroras of Autumn” [a long poem written in 1948, published in a 1950 book of the same name with a collection of 31 other poems, which won the National Book Award for Poetry in 1951]. That poem is about Stevens eyeing the end of his life by the dazzling, otherworldly light of the northern lights. It’s a poem of extraordinary beauty. In this play, that dazzling display of natural wonder is actually AI. It’s no longer the sublime of nature.

Goldberg: Were you picturing Robert as you wrote this character?

Akhtar: I write to an ideal; it’s not necessarily a person.

Robert Downey Jr.: I feel that me and ideal are synonymous.

Akhtar: Robert’s embodiment of McNeal is in some ways much richer than what I wrote.

Downey: I have a really heavy, heavy aller­gy to paper. I’m allergic to things written on paper.

Akhtar: As I’ve discovered!

Downey: But the writing was transcendent. The last time that happened, I was reading Oppenheimer.

Goldberg: There’s Oppenheimer in this, but there’s also Age of Ultron, right?

Downey: Actually, I was thinking about that while I was reading this. And I’ll catch you guys up in the aggregate. I’m only ever doing two things: Either I’m trying to avoid threats or I’m seeking opportunities. This one is the latter. And I was thinking, Why would I be reading this? Because, I mean, I’ve been a bit of an oddball, and I was thinking, Why is this happening to me; why is this play with me? And I’m having this reaction, and it took me right back to Paul Bettany [English actor; b. 1971].

So that you guys understand what’s going on, this is the second Avengers film, Age of Ultron, and Bettany was playing this AI, my personal butler. The butler had gone through these iterations, and [the writer and director] [sic] Joss Whedon [b. 1964] decided, “Let’s have you become a sentient being, a sentient being that is created from AI.” So first Bettany is the voice, and then he became this purple creature. And then there was this day when Bettany had to do a kind of soliloquy that Joss had written for him, as we are all introduced to him, wondering, Is he a threat? Can we trust him? Is he going to destroy us? And there comes this moment when we realize that he’s just seeking to understand, and be understood. And this was the moment in the middle of this genre film when we all stopped and thought, Wait, I think we might actually be talking about something important.

Goldberg: Bart, what are you exploring here?

Bartlett Sher: I’m basically exploring the deep tragedy of the life of Jacob McNeal. That’s the central issue. AI and everything around it, these are delivery systems to that exploration.

Akhtar: Robert has this wonderful moment in the play, the way he does it, in which he’s arguing for art in this very complicated conversation with a former lover. And it gets to one of the essences of the play, which is that this is an attempt to defend art even if it’s made by an indefensible person. Because in the end, human creation is still superior, and none of us is perfect. So the larger conversation around who gets to write, the morality of writing, all of that? In a way, it’s kind of emerging from that.

Goldberg: I can’t say for sure, but I think this is the first play that’s simultaneously about AI and #MeToo [a social movement against sexual harassment and sexual assault launched the U.S. in 2006; popularized as a hashtag in 2017].

Downey: And identity and intergenerational conflict and cancel culture and misunderstanding and subintentional contempt and unconscious bias.

Goldberg: Are there any third rails you don’t touch?

Akhtar: McNeal is the third rail. He’s a vision of the artist in oppo­sition to society. Not a flatterer of the current values, but someone who questions them: “That’s a lie. That’s not true.”

Goldberg: The timing is excellent.

Downey: In movies, you always miss the moment, or you are preempted by something. With Oppenheimer, we happened to be coming out right around the time of certain other world events, but we couldn’t have known. With this, we are literally first to market. Theater is the shortest distance between two points. You have something urgent to say, and you don’t dawdle, and you have a space like Lincoln Center that is not interested in the bottom line, but interested in the form. And you have Ayad inspiring Bart, and then you get me, the bronze medalist. But I’m super fucking motivated, because I never get this sense of immediacy and emergence happening in real time.

Goldberg: Let’s talk for a minute about the AI creative apocalypse, or if it’s a creative apocalypse at all. I prompted Claude [family of LLMs developed by Anthropic in 2023] to write a play just like McNeal, with the same plot turns and characters as your play, and I asked it to write it in your style. What emerged was a play called The Plagiarist’s Lament. I went back and forth with Claude for a while, mainly to try to get something less hackish. But in the end, I failed. What came out was something like an Ayad play, except it was bad, not good.

Akhtar: But here’s the thing. You’re just using an off-the-shelf product, not leading-­edge story technology that is now becoming increasingly common in certain circles.

Goldberg: So don’t worry about today, but tomorrow?

Akhtar: The technology’s moving quickly, so it’s a reality. And worrying? I’m not trying to predict the future. And I’m also certainly not making a claim about whether it’s good or bad. I just want to understand it, because it’s coming.

Downey: To borrow from recent experience, I think we may be at a post-Trinity [first detonation of a nuclear weapon, conducted in New Mexico as part of the Manhattan Project on 16 July 1945], pre-Hiroshima [6 August 1945], pre-Nagasaki [9 August 1945] moment, though some people would say that we’re just at Hiroshima.

Goldberg: Hiroshima being the first real-world use of ChatGPT [popular computer program, based on an LLM and launched by OpenAI in 2022, that holds conversations through a chat room]?

Downey: Trinity showed us that the bomb was purpose-built, and Hiroshima was showing us that the purpose was, possibly, not entirely necessary, but that it also didn’t matter, because, historically, it had already happened.

Goldberg: Right now, I’m assuming that part of the problem I had with the LLM was that I was giving it bad prompts.

Downey: One issue is that LLMs don’t get bored. We’ll be running something and Bart will go, “I’ve seen this before. I’ve done this before.” And then he says, “How can I make this new?”

The people who move culture forward are usually the high-ADD folks that we’ve tended to think either need to be medicated or all go into one line of work. They have a low threshold for boredom. And because they have this low threshold, they say, “I don’t want to do this. Do something different.” And it’s almost just to keep themselves awake. But what a great gift for creativity.

Goldberg: The three of you represent the acting side, and directing, and writing. Who’s in the most existential danger here from AI?

Downey: Anyone but me.

Akhtar: The Screen Actors Guild has dealt with the image-likeness issue in a meaning­ful way.

[The Screen Actors Guild (SAG) was the labor union which represented film and television performers in the United States. In 2012, SAG merged with the American Federation of Television and Radio Artists (AFTRA) to create SAG-AFTRA, which represents approximately 160,000 media professionals in various fields in the U.S. In solidarity with the Writers Guild of America, SAG-AFTRA launched a strike in 2023 precipitated in part by new technologies like AI and digital recreation. See ROT posts “2023 Writers Guild Strike” and “‘The Actors' Strike Dims a Bright Spot For New York City,’” referenced in the afterword to this post.]

Downey: We’ve made the most noise—­we, SAG—­and we’re the most dramatic about everything. I remember when I was doing Chaplin [1992; produced by Carolco Pictures; distributed by TriStar Pictures; Downey portrayed Charlie Chaplin (1889-1977)], the talk was about how significant the end of the silent era was.

Goldberg: Is this the same level of disruption?

Downey: I doubt it, but not because Claude can’t currently pin his ass with both hands. There are versions that are going to be significantly more advanced. But technologies that people have argued would impede art and culture [see several of the posts referenced in the afterword] have often assisted and enhanced. So is this time different? That’s what we’re always worrying about. I live in California, always wondering, Is that little rumble in the kitchen, is this the big one?

Sher: For me, I think directing is very plastic. It requires integrating a lot of different levels of activity. So actually finding a way to process that into a computer’s thinking, and actually having it work in three dimensions in terms of organizing and developing, seems very difficult to me. And I essentially do the work of the interpreter and synthesizer.

A machine can tell you what to do, but it can’t interact and connect and pull together the different strands.

Akhtar: There’s a leadership dimension to what Bart does. I mean, you wouldn’t want a computer doing that.

Sher: This could sound geeky, but what is the distinguishing quality of making art? It is to participate in something uniquely human, something that can’t be done any other way.

So if the Greeks are gathering on the hillside because they are building a space where they can hear their stories and participate in them, that’s a uniquely human experience.

Akhtar: I do think that there is something irreducibly human about the theater, and that probably over time, it is going to continue to demonstrate its value in a world where virtuality is increasingly the norm. The economic problem for the theater has been that it happens only here and only now. So it’s always been hard to monetize.

Goldberg: But I have two words for you: ABBA Voyage. I mean, it’s an extraor­dinarily popular show that uses CGI [computer-generated imagery] and motion capture to give the experience of liveness without ABBA actually being there. Not precisely theater, but it is scalable, seemingly live technology.

[ABBA Voyage is a virtual concert residency by the Swedish pop group which débuted in 2022 and is currently scheduled to run through 2026. It depicts the group as they appeared in 1979 using vocals re-recorded for this show by the group in a studio, accompanied by a live band on stage.]

Downey: Strangely, this is the real trifecta: IP, technology, and taste. I think of this brand of music—which, you know, it’s not my bag, but I still really admired that somebody was passionate about that and then purpose-built the venue. And then they said, “We’re not going to go for ‘Oh my God, that looks so real.’ We’re actually going to go for a more two-dimensional effect that is rendered in a way in which the audience can complete it themselves.”

[‘IP’ is an abbreviation for so many different phrases, several of which are potentially applicable here, that I cannot interpret its use here definitively.  The best known use is for ‘Internet Protocol’ (the set of rules governing the format of data sent via the internet, as in ‘IP address’), but my best guess is that Downey means ‘intellectual property.’]

Akhtar: ABBA Voyage is an exception. But it’s still not live theater.

Sher: It’s also not possible without the ABBA experience that preceded it. It’s an augmentation; it’s not original.

Goldberg: In terms of writing, Ayad, I did what you suggested I do and asked Claude to critique its own writing, and it was actually pretty good at that. I felt like I was actually talking with someone. We were in a dialogue about pacing, clarity, word choice.

Sher: But it has no intuition at all, no intuition for Ayad’s mindset in the middle of this activity, and no understanding of how he’s seeing it.

Downey: It does have context, and context is critical. I think it’s going to start quickly modeling all of those things that we hold dear as ­subtleties that are un­assailable. It’s going to see what’s missing in its sequence, and it’s going to focus all of its cloud-bursting energy on that.

Goldberg: It might be the producers or the studios who are in trouble, because the notes are delivered sequentially, logically, and without defensiveness. Do you think that these technologies can give better notes than the average executive?

Akhtar: I know producers in Hollywood who are already using these tools for their writers. And they’re using them empirically, saying, “This is what I think. Let’s see what the AI thinks.” And it turns out that the AI is actually pretty good at understanding certain forms. If you’ve got a corpus of texts—like, say, Law & Order [television crime drama on NBC-TV, 1990-2010; 2022-Present] ; you’ve got many, many seasons of that, or you’ve got many seasons of a children’s show—those are codified forms. And the AI, if it has all those texts, can understand how words are shaped in that form.

Goldberg: So you could upload a thousand Law & Order scripts and Claude could come up with the thousandth and first.

Akhtar: About a year and a half ago, when I started playing with ChatGPT, the first thing that I started to see were processes of language that reminded me of reading Shakespeare. No writer is better at presenting context than Shakespeare. What I mean by that is Shakespeare sets everything quickly in motion. It’s almost like a chess game—you’ve got pieces, and you want to get them out as quickly as possible so you have options. Shakespeare sets the options out quickly and starts creating variations. So there is a series of words or linguistic tropes for every single play, every poem cycle, every sonnet. They all have their universe of linguistic context that is being deployed and redeployed and redeployed. And it is in that play of language that you find an accretion of meaning. It was not quite as thrilling to see the chatbot [computer program designed to respond with conversational or informational replies to verbal or written messages] do it, but it was actually very interesting to recognize the same process.

Goldberg: Shakespeare was his own AI.

Downey: Because he performed as a younger man, it was all uploaded into Shakespeare’s system. So he was so familiar with the template, and he had all this experience. And similarly, all of these LLMs are in this stage where they are just beginning to be taken seriously. It’s like we’re pre-bar mitzvah, but these are sharp kids.

[For readers who aren’t students of theater history, William Shakespeare (1564-1616) was an established actor from about 1585, years before he wrote his first play. Some of this is discussed in “Shakespeare's Development as a Dramatist” by Kirk Woodward (2 and 5 March 2025).

[The bar mitzvah is the Jewish religious ceremony to admit a boy of 13 to the community as an adult member; the girls’ ceremony, at 12 or 13, is called a bat mitzvah. (Downey described himself in a 2004 New York Times interview as a “Jewish-Buddhist.”)]

Goldberg: Would you use ChatGPT to write an entire piece?

Sher: Soon we’ll be having conversations about whether Claude is a better artist than ChatGPT. Could you imagine people saying, “Well, I’m not going to see that play, because it was written by this machine; I want to see this one, because it’s written by Gemini [chatbot developed in 2023 by Google] instead.”

Goldberg: Unfortunately, I can easily imagine it.

Akhtar: I’m not sure that I would use an LLM to write a play, because they’re just not very good at doing that yet, as you discovered in your own play by Claude. I don’t think they’re good enough to be making the kinds of decisions that go into making a work of art.

Goldberg: But you’re teaching the tool how to get better.

Akhtar: So what? They’ve already gone to school on my body of work.

[Read: The authors whose pirated books are powering generative AI]

Goldberg: So what? So what? Six hundred years of [Johannes] Gutenberg [German; ca. 1393-1406 – 1468], and the printing press [ca. 1440] never made decisions on its own.

Akhtar: But we’re already within this regime where power and monetized scale exist within the hands of very few. We’re doing it every day with our phones; you’re teaching the machine everything about you and your family and your desires. This is the paradigm for the 21st century. All human activity is passing through the hands of very few people and a lot of machines.

Goldberg: McNeal is about lack of control.

Akhtar: It is. I’m just making the point that we’re not really in a different regime of power with AI. It may be even more concentrated and even more consequential, but at the end of the day, to participate in the public space in the 21st century is to participate in this structure. That’s just what it is. We don’t have an alternative, because our government has not regulated this.

Goldberg: You see the LLM as a collaborator in some ways. Where will the red line be for writers, between collaboration and plagiarism?

Akhtar: From my perspective, there are any number of artists we could look at, but the one that I would probably always spend the most time looking at is Shakespeare, and it’s tough to say that he wasn’t copying. As McNeal explains at one point in the play, King Lear shares 70 percent of its words with a previous play called King Leir, which Shakespeare knew well and used to write Lear. And it’s not just Leir. There’s that great scene in Lear where Gloucester is led to this plain and told it’s a cliff over which he’s going to jump, and that subplot is taken right out of Sir Philip Sidney [English poet, courtier, scholar, and soldier; 1554-86]. It may reflect deeper processes of cognition. It may reflect, as Bart has said, how we imitate in order to learn. All of that is just part of what we do. When that gets married to a corporate-ownership model, that is a separate issue, something that will have to get worked out over time, social­ly and legally. Or not, if our legislators don’t have the will to do so.

[Records show that The moste famous Chronicle historye of Leire king of England and his Three Daughters, an anonymous Elizabethan play, was performed on 6 and 8 April 1594 by a cast that combined personnel from Queen Elizabeth’s Men and Sussex’s Men. Shakespeare, who might have been a player in the Queen's company, may have performed in King Leir. Shakespeare’s King Lear is thought to have been composed sometime between 1603 and 1606. The above scene in King Lear [Act 4, Scene 6] is a direct adaptation of a plot device from Sidney’s 1593 novel The Countess of Pembroke’s Arcadia]

Goldberg: The final soliloquy of the play—no spoilers here—is augmented by AI.

Akhtar: This has really been a fascinating collaboration. Because I wanted some part of the play to actually be meaningfully generated by ChatGPT or some large language model—Gemini, Claude. I tried them all. And I wanted to do it because it was part of what the play was about. But the LLMs had a tough time actually delivering the goods until this week. I’ve finally had some experiences now, after many months of working with them, that are bearing fruit.

I wanted the final speech to have a quality of magic to it that resembles the kind of amazement that I knew you had felt working with the model, and that I have sometimes felt when I see the language being generated. I want the audience to have that experience.

Sher: You know, I think the problem you were facing could have been with any of your collaborators. We just had this new collaborator to help with that moment.

Goldberg: You’re blowing my mind.

Akhtar: It’s not really that controversial.

Goldberg: Yes it is. It’s totally controversial.

Downey: Well, let’s find out!

Goldberg: It’s more of a leap than you guys think.

Akhtar: It’s a play about AI. It stands to reason that I was able, over the course of many months, to finally get the AI to give me something that I could use in the play.

Downey: You know what the leap was like? A colicky little baby finally gave us a big ol’ burp.

Akhtar: That’s exactly right. That’s what happened. A lot of unsatisfying work, and then, unprompted, it finally came up with a brilliant final couplet! And that’s what I’m using for the end of the play’s final speech.

Goldberg: Amazing, and threatening.

Sher: I just can’t imagine a world in which ChatGPT could take all experience and unify it with Ayad’s interest in beauty and meaning and his obsession with classical tragedy and pull all those forces together with emotion and feeling. Because no matter how many times you prompted it, you’re still going to get The Pestilential Plagiarist, or whatever it’s called.

Downey: The reason that we’re all sitting here right now is because this motherfucker, Ayad, is so searingly sophisticated, but also on occasion—more than occasionally—hot under the collar. My new favorite cable channel is called Ayad Has Fucking Had It. He’s like the most collaborative superintelligence you will ever come across, and therefore he’s letting all this slack out to everyone around him, but once in a while, if this intelligence is entirely unappreciated for hours or days at a time, he will flare. He’ll just remind us that he can break the sound barrier if he wants to. And I get chills from that. And that’s why we’re here. It’s the human thing.

Akhtar: It’s not new for humans to use tools.

Sher: Are we going to be required to upload a system of ethics into the machines as they get more and more powerful?

Downey: Too late.

Goldberg: That’s what they promise in Silicon Valley, alignment with human values.

Downey: Two years ago was the time to do something.

Akhtar: You guys are thinking big. But I just don’t know how this is going to play out. I don’t know what it is. I’m just interested in what I’m experiencing now and in working with the technology. What’s the experience I’m having now?

Goldberg: There’s a difference between a human hack and an excellent human writer. The human hack doesn’t know that they’re bad.

Downey: This is a harebrained rabbit hole where we could constantly keep thinking of more and more ramifications. Another issue here is that certain great artists do something that most people would labor an entire life or career to come close to, and the second they’re done with it, they have contempt for it, because they go, “Eh, that’s not my best.”

Akhtar: I recognize someone in that.

Downey: All I’m saying is that I just want the feeling of those sparks flying, that new neural pathway being forced. I want to push the limits. It’s that whole thing of pushing limits. When I feel good, when I can tell Bart is kicking me, when Ayad is just lighting up, and when I’m realizing that I just got a note that revolutionized the way I’m going to try to portray something, you go, “Ooh!” And even if it’s old news to someone else, for me, it’s revolutionary.

Akhtar: Another way of putting this, what Robert is saying, is that what he’s engaged in is not problem-solving, per se. It’s not that there’s an identified problem that he is trying to solve. This is how a computer is often thinking, with a gamification sort of mindset. For Robert, there’s a richness of the present for him as he’s working that is identifying possibilities, not problems.

Sher: I’ve thought a lot about this, trying to understand the issue of GPT [‘generative pre-trained transformer’ (GPT) is a type of LLM and a framework for generative artificial intelligence] and creativity, and I’m a lot less worried now, because I feel that the depth of the artistic process in the theater isn’t replicable.

The amalgam of human experience and emotion and feeling that passes through artists is uniquely human and not capturable. Word orders can be taken from all kinds of sources. They can be imitated; they can be replicated; they can be reproduced in different ways. But the essential activity of what we do here in this way, and what we build, has never been safer.

Downey: And if our job is to hold the mirror up to nature, this is now part of nature. It is now part of the firmament. Nature is now inclusive of this. We’re onstage and we’re reflecting this back to you. What do you see? Do you see yourself within this picture?

[The theater website Broadway World did a “Review Roundup” of McNeal and, based on 20 notices, it calculated an average rating of 41%.  Of the 20 reviews samples, 2 were positive (˄ – 10%), 5 were mediocre (˃ – 25%), and 13 were negative (˅ – 65%).  Here’s a selection of the excerpts BWW posted in the roundup:

˅ Jesse Green, New York Times:  I’m afraid, alas, the pixel wins, because the play, which opened on Monday, in a stylish Lincoln Center Theater production directed by Bartlett Sher, works only as provocation.  Timely but turgid, it rarely rises to drama; in a neat recapitulation of current fears about technology, its humans, hardly credible as such, have been almost entirely replaced by ideas. 

˃ Charles Isherwood, Wall Street Journal: “McNeal,” directed by Lincoln Center Theater’s newly named executive producer Bartlett Sher, is itself a confused and discursive if thought-provoking drama that often seems a grab bag of ideas Mr. Akhtar delves into without finding much depth.

˅  Aramide Timubu, Variety: “McNeal” falters because it doesn’t know what it wants to say. Moreover, the narrative felt confusing and meaningless, with a mix of genres and no actual theme or climax. Jacob is a writer, but viewers can never penetrate who he is beyond the surface. Instead, the play becomes an illustration of a self-absorbed man who treats the people around him like accessories for his own gain.

˃ Jeremy Gerard, Deadline: With much 21st Century razzle dazzle provided by the magnificent video projection designs of Jake Barton and huge digital composite images – video AI projections of faces of the actors meld into one another at one point – from AGBO, McNeal, astutely directed by the great Bartlett Sher, is an often confusing though wheedlingly emotional mindgame. The discombobulation is, one suspects, Akhtar’s intention, a way of presenting on a physical stage a near-future realm of thorny, magic-seeming complexity in which thousands of years of data – from Shakespeare and Ibsen to your dead wife’s old notebooks, and everything in between – can be melded into a book with your name on it, and in relative minutes. Is this theft, or merely a literary Moog waiting for its Brian Eno?

˅ Johnny Oleksinki, New York Post: Should your sole aim be to watch the Marvel and “Oppenheimer” actor, who’s making his Broadway debut, give a capable performance in his signature Tony Stark staccato, mission accomplished. However, it is, well, a marvel how even the most blinding star power can dim when blacked out by a mind-numbing plot, mouthpiece supporting characters and a Universal Studios-scale set of giant screens that’s an expensive apology for the actual play.

˅ Jackson McHenry, Vulture/New York: McNeal’s woozy ruminations about art and technology might strike with more force if the actual drama around them had more tensile strength. The human dynamics Akhtar and Sher hang all of this on never get past cliché: McNeal confronts a cadre of women, including Martin’s assistant (Saisha Talwar), a horrifyingly underprepared magazine journalist (Brittany Bellizeare), and a former New York Times books editor who is pretty obviously based on Pamela Paul (Melora Hardin). In his exchanges with them, they get to do little except absorb his rants about everything from the work of Annie Ernaux to Harvey Weinstein. If McNeal wants to tell us that a great artist — specifically a great man, in this case — is some unique force in the universe, it requires a more finely crafted rendering of that man. But on the opposite end, if the play wants to suggest AI could offer an alternative to that figure that’s equal in stature? Well, there’s an issue of rendering there, too.

˃ Emlyn Travis, Entertainment Weekly: McNeal might ask fascinating questions about a writer’s sense of integrity, the disconnect between generations of writers, and the limits of artificial intelligence within art, but the delivery is so garbled that it is difficult to ascertain what the play is actually trying to say about these complex issues… if anything at all. Grade: C+

˄ Frank Scheck, New York Stage Review: Star power is not to be underestimated. Not only can it bring in audiences who may not normally attend new Broadway plays, it can also infuse a problematic work with a gravitas it might not normally possess. Such is the case with Lincoln Center Theater’s production of the new drama by Ayad Akhtar (Disgraced, The Invisible Hand) that touches on such themes as artificial intelligence, plagiarism, and writerly inspiration without really coming to grips with any of them. But it doesn’t matter, thanks to the presence of Robert Downey Jr. in the title role. Downey, who recently won an Oscar for the film Oppenheimer and is here making his stage debut, brings such charisma and magnetism to McNeal that it’s easy to overlook the play’s flaws.

˄ Roma Torre, New York Stage Review: Before I get into the merits of the play and Lincoln Center’s thrilling production, something else that’s very real is the extraordinary talents of its leading man. Making his Broadway debut, Robert Downey Jr., who most of us know only through his film work, is a great stage actor. As famed novelist Jacob McNeal, an egotistical, self-acknowledged asshole, he is a fascinating creature treading the line between obnoxious and seductive. You can’t take your eyes off him . . . and happily, he’s on stage for the entirety of Akhtar’s 90-minute drama.

˃ Joe Dziemianowicz, New York Theatre Guide: In Ayad Akhtar’s crisply staged yet dramatically muddy play, McNeal, Oscar winner Robert Downey Jr. summons every ounce of his innate swagger and smugness as a misogynistic author who’s just found himself the recipient of the Nobel Prize for Literature. In his Broadway debut, Downey steps up; he’s sure-footed and magnetic. The play itself, less so.

[I have published (or republished) several articles concerning computers or artificial intelligence and theater or the arts in general.  They aren’t all relevant to “The Playwright in the Age of AI,” but they all do relate to the same basic phenomenon.  For that reason, I’m appending a list of those ROT posts for your reference:

   •   “Theater and Computers” (5 December 2009)
•   “Computers and Actors” (4 and 7 October 2021)
•   “‘Entertainment in the Age of AI’” (22 August 2022)
•   “‘Use of artificial intelligence generates questions about the future of art’” by Jeffrey Brown and Alison Thoet (7 March 2023)
•   “‘AI in the Arts Is the Destruction of the Film Industry. We Can't Go Quietly’” by Justine Bateman (4 June 2023)
•   “2023 Writers Guild Strike” (1 June 2023)
•   “‘The Actors' Strike Dims a Bright Spot For New York City’” by Stefanos Chen (25 July 2023)

[Jeffrey Goldberg (b. 1965) is the editor-in-chief of The Atlantic and the moderator of Washington Week With The Atlantic on the Public Broadcasting Service (PBS).]