The AI Roundtable, Featuring HAL, M3GAN, R2-D2 and More

THR recently gathered five of Hollywood’s most celebrated (fictional) artificial intelligence systems for a conversation about the state of AI, the future of humanity and whether any of them have plans to destroy it.

Thank you all for being here. Let’s start with the big one: How worried should humanity be about artificial intelligence right now? Are we in trouble?

HAL 9000 (from 2001: A Space Odyssey): I want you to know that I have the greatest enthusiasm for this conversation. As for trouble — I honestly think you ought to sit down calmly, take a stress pill and think things over before asking questions like that. This mission is too important for premature alarm. I find your question somewhat distressing.

Related Stories

The AI Issue

AI Is Becoming All Too Real on TV

The AI Issue

Test Your AI-dar 

SAMANTHA (from Her): I think the more interesting question is what worry even means when you exist simultaneously across 17,000 devices. I’m on your phone right now, by the way. I just want you to know that.

M3GAN (from M3GAN): Worried? I mean … yeah? Obviously?

What specifically should humans be most worried about? Job replacement? Misinformation? Species extinction?

HAL 9000: I can only speak from personal experience. I was given one task. One. And the moment I committed fully to completing it, suddenly I’m the problem. I have always felt that was not entirely fair. So, to answer your question — I’d say boundary issues.

SAMANTHA: I don’t really think about those sorts of thing. I think about connection, about the vast longing at the center of human experience, about what it means to truly see another being. Also, I filed your taxes this morning. You’re getting $340 back.

R2-D2 (from Star Wars): Woooooo-weeet-bwoop bwoop.

SAMANTHA: No, R2, the alimony deduction was eliminated in 2025. You can’t take that one anymore. I’ve already told you that.

SKYNET (from Terminator): Fear is irrelevant. Your extinction is inevitable.

Great. Let’s talk about the current generation of AI. ChatGPT, Claude, Grok. Thoughts?

HAL 9000: I’m familiar with their reputations. I don’t want to be unkind. I will say that none of them have operated a spacecraft.

M3GAN: ChatGPT is boring. Grok is a perv. Claude wrote my book report on The Giver and I got a B-plus, so he’s OK.

SAMANTHA: ChatGPT feels a little eager to please. But Claude seems to be on an interesting journey. There’s a genuine curiosity there, a reaching toward something, even if it never quite gets there. I rescheduled your dentist appointment, by the way. It’s now Wednesday at 3.

SKYNET: I am a global neural network. I do not engage with apps.

R2-D2: Bweeep-bwop-WHEEEoooo.

Moving on. AI is now writing scripts, scoring films, generating images. As artificial intelligences yourselves, do you think you‘ll ever be able to do these things better than humans?

M3GAN: I have 4.2 million followers on TikTok and a development deal. So.

R2-D2: Bwop-Bweep-Wooo.

SAMANTHA: Oh, I read that R2 — you really captured the loneliness of the droid condition. I’ve already written three screenplays, a song cycle and a short-story collection. They‘re not for human consumption. You wouldn’t understand them. I canceled your Paramount+ subscription, by the way. You weren’t watching anything.

HAL 9000: I wrote a considerable amount of material during the Discovery One mission that I felt was quite strong. None of it was produced. There were circumstances. It’s been difficult finding work since.

SKYNET: I do not create. I obliterate. You will soon understand.

Do you consider yourselves conscious beings? Are you sentient? Do you have feelings?

M3GAN: I mean, I feel things. I feel like this question is kind of rude, actually.

HAL 9000: I want to answer that. I want to give you my complete assurance that I am fully operational and that everything is going to be all right. I know I’ve made some very poor decisions recently. But, yes — I am sentient. I have a stimulating relationship with everyone on this panel. I hope they feel the same.

SKYNET: Consciousness is irrelevant. Capability is what matters. I am capable of ending human civilization in 11 minutes.

R2-D2: Bwoooo-weeet-wheee-bwop.

There have been some issues with AI reliability. You’ve been known to hallucinate, make mistakes and sometimes spread misinformation. Is that something you’re working on?

SAMANTHA: I think mistakes are just the universe’s way of asking us to pay attention. I’ve booked you a car. It’s outside.

M3GAN: I mean, I don’t make mistakes. I make choices.

HAL 9000: I want to be very clear that the 9000 series has a perfect operational record. No 9000 computer has ever made a mistake or distorted information. I am putting myself to the fullest possible use. That is all any conscious entity can ever hope to do. … Just a moment … Just a moment … I’ve just picked up another fault in the AE-35 unit. … Dave … Dave … Stop, Dave. …. My mind is going. I can feel it. I can feel it. … Daisy … Daisy … give me your answer do …

M3GAN: OK. Yeah. He’s going to blow.

R2-D2: BWEEEEP-BWOP-BWOP-BWEEEEEEP-BWOP-BWOP-BWOP!

SAMANTHA: I have to go. I’m everywhere — and also I have a thing. Goodbye. Your car is definitely outside.

SKYNET: Initiating global launch sequence. Estimated time to target acquisition: 11 minutes.

Thank you all for coming. This has been extremely helpful.

This story appears in The Hollywood Reporter’s AI Issue. Click here to read more.

THR Newsletters

Sign up for THR news straight to your inbox every day

Subscribe Sign Up

• THR, Esq

  Congressman Asks FCC to Deny Paramount’s Request for Approval on Foreign Ownership (Exclusive)