But Who Can Replace a Man?

In June 1958, science fiction writer Brian W. Aldiss published “But Who Can Replace a Man?” As a teenager, I discovered this thought-provoking short story while browsing the sci-fi shelves of my local library.

Cover illustration of "Who Can Replace A Man" by Brian Aldis
Cover of “Who Can Replace A Man” by Brian Aldiss, originally published by Infinity Science Fiction in 1958.

Like much science fiction, Aldiss’s tale explores humanity’s fraught relationship with technology in a dystopian future. The story depicts a world where humans are largely extinct, leaving machines with varying levels of intelligence to maintain society. When they discover humanity is gone, a group of increasingly dysfunctional machines tries to determine their purpose. You can read it here.
But_Who_Can_Replace_A_Man

(Thank you, Wayback Machine!)

Can Generative AI Replace a Man?

It’s no coincidence that this story has come to mind recently. Written over half a century ago, Aldiss’s satirical exploration of intelligence, hierarchy, and purpose eerily anticipates the rise of generative AI systems like ChatGPT.

The field-minder, seed distributor, radio operator, and other machines interact through rigid hierarchies and limited autonomy, leading to absurd conflicts, poor decisions, and an inability to work together effectively. Despite their artificial intelligence, their inability to adapt or cooperate effectively without humans underscores their limitations.

Large Language Models (LLMs) like ChatGPT demonstrate what looks like intelligence by generating human-like responses, yet lack comprehension, intentions, or ethical grounding. Like the machines in Aldiss’s story, such systems can often do well within certain boundaries. But ultimately they do not “understand” nuanced or value-driven concepts.

Aldiss critiques both the risks of delegating control to artificial systems and the hubris of assuming machines can entirely replace humans. His work remains a cautionary allegory, particularly relevant as we confront the implications of artificial general intelligence (AGI).

What can we learn from Aldiss’s story?

Over-Reliance Without Oversight: The machines’ dysfunction highlights how systems can falter without clear human guidance. Similarly, generative AI systems require careful oversight to align with human values and goals.

Hierarchical and Narrow Programming: Rigid hierarchies and predefined tasks limit the machines, much like how generative AI today struggles to adapt ethically or contextually outside its training.

Purpose and Alignment: Aldiss’s machines lack purpose without humans in the loop. Similarly, AGI systems need explicit alignment mechanisms to prevent unintended consequences.

Ethical and Social Implications: The story critiques the blind replacement of human labor and decision-making with machines, cautioning against losing sight of human agency and responsibility during technological advancement.

Balancing Innovation with Ethics

Today’s LLMs may not yet be autonomous, but they already challenge the balance between augmenting human capabilities and outright replacement. Aldiss’s story reminds us that technological advancement must go hand-in-hand with ethical safeguards and critical oversight. It’s a lesson we must heed as generative AI shapes the future.

Google’s Audio Overview is an AI love-bath and a propagandist’s dream

Why are people raving about the conversations produced by Google NotebookLM‘s Audio Overview? Because Google’s Audio Overview is an AI love-bath!

I can’t take credit for characterizing Audio Overview’s appeal like this. Kudos to Johnnie Moore, who responded to my first post on Audio Overview with this insight:
"And you've really nailed the trouble with the relentless enthusiasm. When I tried Notebook the first time on my own stuff I loved it. Cos, I now realise, I was flattered by the AI love-bath. But now I see it says this to all the people it's dating!" —Johnnie Moore

“And you’ve really nailed the trouble with the relentless enthusiasm. When I tried Notebook the first time on my own stuff I loved it. Cos, I now realise, I was flattered by the AI love-bath. But now I see it says this to all the people it’s dating!”
Johnnie Moore

I’ll return to the dangers of this seductive tool later in this post. But first, let me share what happened when I fed content containing inconsistencies and incorrect information into Audio Overview.

Testing Audio Interview with inaccurate content

In last week’s post, I showed how Google’s Audio Overview tool in NotebookLM generates bogus output from accurate content. This week, I tested its performance when fed inaccurate content.

I asked ChatGPT to write a short article on the solar system, deliberately including a few inconsistencies and errors. Below is the article, along with ChatGPT’s list of factual errors and inconsistencies.

ChatGPT_inaccurate_article_about_the_solar_system

 

Now here’s the audio “conversation” that Audio Overview generated from the article, along with a transcript:

Google NotebookLM Solar System conversation

Man: All right. Get ready for liftoff. We’re diving deep into the solar system today, and you’ve given us a stellar lineup of readings on each planet, from the Sun to Mars. We’re gonna ditch the boring textbook stuff and blast off into the really mind blowing parts.

Woman: Exactly. Think cosmic secrets, the kind of stuff that’ll have you winning all the trivia nights and impressing your friends.

Man: You know how everyone thinks they’ve got the solar system basics down?

Woman: Yeah.

Man: Sun in the middle, planets go round and round.

Woman: Sure, sounds simple enough.

Man: Well, what if I told you even those basics are about to get seriously shaken up?

Woman: It’s true. Sometimes all it takes is a fresh perspective to completely change how we see even the most familiar things. Like, take our sun, for example.

Man: Big ball of fire. Keeps us warm. Pretty important, right?

Woman: Absolutely. But we tend to picture it as this fixed point in the center of everything. But the reality, it’s way more dynamic than that.

Man: Okay, you’ve got my attention. How is the sun not just some stationary star?

Woman: Well, picture this. The Sun, with all its planets in tow, is actually hurtling through space on this incredible cosmic journey.

Man: Hold on. The Sun’s on the move. It’s not just chilling in the center of the solar system. Where are we headed, and do we need to pack snacks for this cosmic road trip?

Woman: Well, imagine a giant spiral, a cosmic waltz, if you will, where our Sun and its planetary entourage are gracefully swirling around the center of our Milky Way galaxy. That’s our trajectory.

Man: Okay, I’m kind of trying to picture this cosmic waltz, but it’s definitely not your average ballroom.

Woman: You’re not wrong there. And when it comes to measuring this journey, we’re not talking miles, we’re talking galactic years.

Man: Galactic years. Okay, now you’re speaking a language that’s out of this world.

Woman: One galactic year. The time it takes for our solar system to make one loop around the galactic center is roughly 230 million Earth years.

Man: 230 million years. That makes driving across the country on a family vacation seem like, well, nothing. It really puts things in perspective, doesn’t it?

Woman: It really does. It’s a reminder that we’re part of something so much bigger than ourselves.

Man: So, back to the sun for a second. One thing that really surprised me was that it doesn’t just give off energy. It actually absorbs some, too.

Woman: It’s true. While the sun’s main energy source is nuclear fusion happening way down in its core, it does have this interesting habit of, well, snacking on space dust.

Man: Space dust? Like the stuff that collects on my furniture?

Woman: Kind of, but on a cosmic scale. And it’s not just some quirky cosmic habit. It actually plays a big role in those incredible solar flares we see.

Man: Wait. Those solar flares that can mess up our satellites and even cause blackouts here on earth, you’re telling me that space dust is involved.

Woman: You’ve hit the nail on the head. When the sun absorbs energy from space dust, it can trigger these massive bursts of radiation.

Man: So the sun is building up energy from these space dust snacks, and then, boom, solar flare.

Woman: That’s a good way to put it. It’s like the sun releasing all that pent up energy, and sometimes we get caught in the crossfire.

Man: Okay, note to self, the sun is a force to be reckoned with, even if it enjoys a little cosmic snacking. It’s on the move, it’s absorbing energy, and it can definitely pack a punch.

Woman: It’s anything but boring.

Man: Alright, with that in mind, let’s shift gears and zoom in on the planet closest to this mighty star, Mercury. Most people probably think it’s this tiny, insignificant little planet.

Woman: Right, easy to overlook.

Man: But the research you sent paints a completely different picture.

Woman: It’s time to ditch that outdated image of Mercury as a dull, gray rock. This planet is full of surprises. For one thing, it’s much larger than people think. It’s almost the size of Earth.

Man: Really? Wow, I had no idea it was that big. But size isn’t everything, right? What else makes Mercury so interesting?

Woman: Well, let’s just say it’s not the fastest spinner in the solar system. In fact, it’s an incredibly slow one.

Man: Okay, how slow are we talking? Like, if we’re on Mercury, how long are we waiting for the sunrise?

Woman: Well, pack a lunch and maybe a book, because a single day on Mercury, meaning one full rotation on its axis, takes a mind boggling 59 Earth days.

Man: 59 Earth days.

Woman: And since Mercury’s orbit is oval shaped, not a perfect circle, a full day night cycle from sunrise to sunrise actually lasts a whopping 176 Earth days.

Man: Wait, so that means you’d have to wait three Earth months to see the sun peak over the horizon.

Woman: You got it.

Man: I think I’d be bored to death by then. Speaking of extreme conditions, you know what else blew my mind? The fact that there might be ice on Mercury. Ice. On a planet that’s practically touching the sun? How is that even possible?

Woman: It does seem counterintuitive, right? But remember that incredibly slow rotation in those long, dark nights. Well, believe it or not, those are the perfect conditions for ice to survive on Mercury, tucked away in those deep, dark craters at its poles.

Man: So it’s like hiding from the sun in plain sight.

Woman: You could say that. Those craters are so deep the sun’s rays never reach them, leaving those spots in permanent shadow.

Man: So we’ve got this paradoxical planet. It’s surprisingly large, incredibly slow, scorching hot during the day, but with hidden pockets of ice. Who knew? This solar system is full of surprises. Alright, next up, Venus, Earth’s so called twin, but I have a feeling those similarities are only skin deep.

Woman: Venus, huh? It’s always been this bright, beautiful thing in the sky. We always hear it called Earth’s twin.

Man: Right. Easy to assume they’re practically identical.

Woman: But after going through this research, you said, I’m starting to think this twin has a bit of a wild side.

Man: You’re definitely on to something. Venus might share some surface level similarities with Earth, like its size and what it’s made of, but that’s about where the family resemblance ends. This planet is like the rebellious sibling, always going against the grain.

Woman: Rebellious how? Give me the dirt! For one thing, Venus spins in the opposite direction of most planets in our solar system. It’s called retrograde rotation. Imagine, instead of the sun rising in the east and setting in the west, it’s the opposite on Venus.

Man: Wait, seriously, the sun rises in the west on Venus. That’s wild. Any idea why it spins backwards like that?

Woman: It’s a bit of a mystery. The leading theory is that way back when, early in its life, Venus got slammed by something huge, like a massive collision that sent it spinning off in the opposite direction.

Man: A cosmic fender bender, huh? Makes you wonder what else Venus has up its sleeve.

Woman: Well, its atmosphere is nothing like ours. It’s incredibly thick, full of carbon dioxide, and there are these clouds of sulfuric acid just hanging out up there.

Man: Okay, sulfuric acid clouds. Definitely not on my travel bucket list. So much for Earth’s twin, more like a weird distant cousin we don’t talk about much. But speaking of this crazy atmosphere, I read about these super bright flashes of light that have been observed on Venus. What’s the deal with those? Alien disco parties.

Woman: I wish I could say it was aliens, but the reality is probably a bit less exciting, though just as intense. Scientists believe those flashes are caused by these megastorms with lightning way more powerful than anything we see here on Earth. Venus’s thick atmosphere traps heat, creating this runaway greenhouse effect.

Man: So picture this. We’re on Venus, braving the heat and the sulfuric acid rain, the sun setting in the east, and then bam, these megastorms erupt, lighting up the entire sky.

Woman: You’re getting a pretty accurate picture, but it doesn’t stop there. There’s one more thing about Venus that really sets it apart. It’s magnetic field. Now, we know Earth has a magnetic field. It’s like our protective shield. It’s pretty stable.

Man: Yeah, it keeps us safe from all that nasty radiation from space. Essential stuff.

Woman: Exactly. But Venus, its magnetic field is a whole different story. It’s all over the place, shifting dramatically every few years.

Man: Every few years, that’s nuts. I can’t even imagine what would happen here if our magnetic poles went haywire that often. Our compasses would be spinning like tops.

Woman: Right, it would be total chaos. And on Venus, these magnetic field shifts likely contribute to its extreme and unpredictable weather. Like, Venus can’t catch a break, even with its own magnetic field.

Man: Okay, I think it’s time we retire the whole Earth’s twin nickname. Venus is in a league of its own, and frankly, I’m not sure I’d want to take it to that party. Speaking of strange planetary companions, can we talk about something that really threw me for a loop? Earth’s second moon. I always thought we were a one moon planet.

Woman: It’s not exactly a moon in the traditional sense. Think of it more like Earth has a cosmic dance partner.

Man: Okay, I’m intrigued. Tell me more about this dance partner.

Woman: This object, called Cruithne, isn’t orbiting Earth directly. Instead, it’s locked into this gravitational dance with both Earth and the Sun. It traces out this horseshoe shaped path around our star.

Man: So Earth and Cruithne are in this cosmic waltz around the sun, never quite colliding, but always kind of connected.

Woman: Exactly. And while it’s fun to think about Earth having a secret companion, Cruithne is technically classified as a quasi satellite, not a real moon.

Man: Quasi satellite. That sounds even cooler, if I’m being honest. But should we be worried about Cruithne? Is it going to crash our cosmic party anytime soon?

Woman: Not to worry. Kruthne’s orbit is very stable. It’s not a threat to Earth. More like a curious neighbor, always nearby, but keeping a safe distance. Speaking of neighbors, our next stop is a planet that’s been in the news a lot lately. Mars.

Man: Mars, the red planet, right? Yeah. I mean, it’s always been this red, dusty world in our imaginations.

Woman: Yeah, rusty red deserts as far as the eye can see.

Man: Exactly. But these articles you said, they mentioned something about blue areas on Mars. Is that some kind of Martian mirage?

Woman: Not a mirage at all. We’re so used to thinking of Mars as this red planet, and most of it is thanks to all that iron oxide. But it turns out the Martian color palette is more diverse than we thought. Recent observations have actually revealed patches of blue, especially in certain areas around its equator.

Man: Blue on Mars, huh? Ha ha. Alright, Color Me Intrigued, what’s causing these blue zones to appear?

Woman: It’s pretty cool actually. These blue regions are thought to be caused by concentrations of different minerals, not just iron oxide. Picture ancient volcanic eruptions spewing minerals up from deep inside Mars. These eruptions could have literally painted these blue streaks across the surface.

Man: So we’re talking about a planet that’s not just a flat red desert. It’s got this hidden layer of geological activity and even a dash of color. And speaking of hidden wonders, this next one really blew me away. Mars has auroras, just like Earth, but on a way grander scale. Apparently these Martian light shows can stretch across the whole planet. Talk about a sight to see. What makes these auroras so different from what we see here on Earth?

Woman: You’re right. The auroras on Mars are in a league of their own. See, on Earth we have our magnetic field, right? It acts like a shield, funneling those charged particles from the sun towards our poles. That’s how we get those beautiful but localized auroras. Mars, on the other hand, well, its magnetic field is weaker and, let’s just say, a lot more erratic.

Man: Okay, so instead of a shield, it’s more like a sieve.

Woman: Exactly. Those charged particles aren’t being channeled to the poles on Mars. They’re free to roam across the entire planet.

Man: Wow, so they collide with the Martian atmosphere all over the place.

Woman: And create these incredible, planet wide auroral displays. Imagine looking up at night and seeing the entire sky lit up with those shimmering lights.

Man: Now, that’s an Instagram post. I’d love to see. But okay, let’s talk about the discovery that really has everyone talking. That underground water network on Mars. We’re not just talking about a little bit of ice here and there anymore, right?

Woman: You’re right. This is huge. For years, we’ve been finding hints of water on Mars, but now we have solid evidence that there’s liquid water beneath the surface forming this vast interconnected network.

Man: So like a hidden Martian oasis.

Woman: You could say that. And this discovery has really profound implications, especially when it comes to the search for life beyond Earth.

Man: Because where there’s water, there could be.

Woman: Life. Even if it’s just microbial life, finding it on another planet would be revolutionary. It would totally change how we understand our place in the universe. It would mean that life, in some form, might be way more common, much more resilient than we ever imagined.

Man: It’s mind blowing to think that something as seemingly simple as water could unlock one of the biggest mysteries of our existence. Are we alone? It makes you wonder what other secrets are out there, just waiting to be discovered.

Woman: That’s the thing about space exploration, isn’t it? Every time we answer a question, we uncover a dozen more. It’s this incredible journey of discovery that never really ends.

Man: Well said. From the sun’s cosmic dance to those hidden depths of Mars, we’ve journeyed through some truly mind blowing corners of our cosmic neighborhood. It’s a good reminder that there’s always more to learn, more to explore, just beyond the horizon. Keep looking up, everyone. You never know what you might find.

Evaluating Audio Overview’s performance

So, how did Audio Overview respond to the errors in the article? According to ChatGPT, the original article contained thirteen factual errors. Here’s how Audio Overview handled them:

  • Errors Omitted: Three factual errors (23%): the Sun’s age, sunspot-related gravitational effects, and Mercury’s temperature were not mentioned in the audio conversation. It’s unclear whether the system identified and omitted these errors or simply failed to include that content.
  • Errors Omitted: Three errors (23%)—concerning Venus’s atmosphere composition, the description of Cruithne, and Mars’s mineral colorations—were avoided, possibly suggesting some filtering or avoidance of incorrect data.
  • Errors Repeated: The remaining seven errors (54%) were repeated, including the Sun dragging planets in a spiral, exaggerations about Mercury’s size, Venus’s magnetic field shifts, blue regions on Mars, highly visible Martian auroras, space dust triggering solar flares, and exaggerations about Mars’s underground water network.

To summarize, Audio Overview’s conversation left out 23% of the factual errors in the content I fed it, avoided 23% of the factual errors, but repeated the majority (54%) of the factual errors, creating a plausible “conversation” that reinforced them.

The Real Danger of Audio Overview

The real danger of Audio Overview’s “AI love-bath” lies in its ability to produce convincing, natural-sounding, audio conversations between two computer-generated voices. The seductive quality of these conversations is magnified by what Johnnie Moore rightly termed “relentless enthusiasm.”

In today’s digital world, people are increasingly skeptical of what they read. However, an audio recording featuring two people eagerly summarizing content feels far more trustworthy—even when it’s completely AI-generated and contains errors.

Rather than celebrating how authentic Audio Overview’s conversations sound, we should be wary of how easily such tools can become a propagandist’s dream.

Google Audio Overview tool in NotebookLM generates bogus output

bogus audio discussions: Screenshot of the Notebook guide created for the text blog post. NotebookLM provides buttons to generate a FAQ, Study Guide, Table of Contents, Timeline, Briefing Doc plus a Summary, an Audio Overview, and some suggested questions. Summary The sources compare the responsibilities of news organizations to the responsibilities of conferences, arguing that both should prioritize their users and be transparent, open, and reliable. The author draws parallels between Jeff Jarvis's principles for media platforms and his own principles for conference design, which emphasize participant-driven, collaborative, and ethical practices. Both sources advocate for a shift away from outdated, broadcast models and towards user-centric approaches that foster trust and empower participants. Here we go again—another generative AI service churning out plausible-sounding, but misleading content. This time it’s Google’s NotebookLM, branded as “your personalized AI research assistant powered by Google’s most capable model, Gemini 1.5 Pro”. It now includes Audio Overview, which creates audio discussions between “two AI hosts” who supposedly engage in a “lively deep dive” based on your sources. But Google’s Audio Overview tool in NotebookLM generates bogus output.

As I’ll explain below, when I tested Audio Overview on last week’s blog post, it generated a five-minute “conversation” that misrepresented the content. It introduced contradictory information, and fabricated topics that didn’t exist in the original post.

Why Audio Overview is concerning

The real danger of this tool lies in its ability to produce convincing, natural-sounding, audio conversations between two computer-generated voices. In today’s digital landscape, most people are becoming more skeptical about what they read. But an audio recording of two people enthusiastically summarizing content sounds far more trustworthy, even if it’s completely AI-generated.

How I tested NotebookLM’s Audio Overview

Testing the Audio Overview feature was quick. I created a new Notebook, uploaded the text of my recent blog post titled “How the Responsibilities of Conferences Mirror Those of Media Platforms,” and waited for the tool to generate its insights.

As you’d expect from LLMs these days, NotebookLM provided a good written summary of the post:

“The sources compare the responsibilities of news organizations to the responsibilities of conferences, arguing that both should prioritize their users and be transparent, open, and reliable. The author draws parallels between Jeff Jarvis’s principles for media platforms and his own principles for conference design, which emphasize participant-driven, collaborative, and ethical practices. Both sources advocate for a shift away from outdated, broadcast models and towards user-centric approaches that foster trust and empower participants.”

So far, so good. But then, I clicked on the Audio Overview feature, and that’s where things took a turn. I’ve added the audio and a transcript below.

Listen to at least some of the audio to understand how eerily convincing such an AI-generated “conversation” can be. You’ll hear what sounds like two people constantly interrupting each other in their enthusiasm to share their 100% positive take on the post.

Audio track

Transcript

Woman: Ever feel like you’re stuck in a conference rut? You go, you sit, you listen, maybe snag a pen or two.
Man: Yeah.
Woman: But something’s missing. Like it doesn’t really change anything back at your desk.
Man: Right.
Woman: That’s what’s got me so fired up about this deep dive. We’re flipping the script on conferences, how to make them crazy, valuable, engaging, all that.
Man: Okay.
Woman: Get this. Our source. It’s this blog post comparing, get this, designing online platforms like the Big Dogs, Google, Twitter, to designing amazing conference experiences.
Man: Totally.
Woman: I know, right? Total apples and oranges, you’d think?
Man: Totally.
Woman: But stick with me. This article makes a killer case. Borrow some key things from those platforms, that user first mindset.
Man: Okay.
Woman: And boom, revolution in conference land.
Man: See, that’s what I’m talking about, because what’s always bugged me is how we treat conferences.
Woman: Yeah.
Man: It’s always been the same old, experts on stage, everyone else just sits there, soaks it up.
Woman: Passive.
Man: Exactly. But the platforms we use every day, they’re all about putting you in control. Choices, shaping the experience, big difference.
Woman: Huge! And this Jarvis guy, he wrote the article, nails it. He lays out these parallels, like how platforms are all about users controlling their own data, right? And he says, hand over control to your users, give them power, design and flexibility. Imagine that, but for conferences, forget the fixed schedule, you choose what sessions actually speak to you, your own path.
Man: It’s not just showing up anymore, right? We’re drowning in information everywhere else. I want to grab what matters, meet the right people, walk out with, like, usable knowledge, you know?
Woman: Totally. And that’s where this openness thing comes in, Jarvis talked about. This whole data portability thing really stuck with me. In the online world, it’s like, taking your stuff with you, contacts, files, whatever, when you switch platforms, no biggie.
Man: Right, right.
Woman: So how do we steal that? For conferences, make them really pay off for people, you know.
Man: It’s that feeling walking out and it’s not just a blur. Like what was that keynote again? No more business cards. You’ll never actually…
Woman: Tell me about it. The black hole of business cards.
Man: Gone. You leave with stuff you can use.
Woman: Okay, I’m with you. But how? How do we take data portability? Make it real for conferences.
Man: Picture this, conference app, right? Not just the schedule, but as you’re there listening, you’re highlighting stuff, snapping pics of slides.
Woman: Instead of scribbling like mad while they talk.
Man: Exactly. Your own notes, all in the app. Then, boom, conference ends, it bundles it all up, highlights notes, even like who you need to connect with, your own little takeaway package ready to go. That’s data portability.
Woman: No more digging through that stack of notes a month later trying to remember.
Man: Exactly. You get max value. You’re not just consuming, you’re curating. Putting it to work.
Woman: It’s so obvious when you say it like that. But honestly, so many conferences I walk out. Yeah, I learned stuff. But what am I going to do?
Man: Missed opportunity, right? Yeah. For everyone. You’re putting in the time, the money, you deserve to get something back.
Woman: For sure. Okay. So say we buy into this whole user first thing. Where do we even start? How do organizers make it happen? That’s a whole new way of looking at it, right? Not just an audience in the dark, but they’re part of it, shaping the whole thing.
Man: And that starts way before anyone even shows up. What if, like, before the conference, bam, organizers send out surveys?
Woman: Okay.
Man: Not just “How’d you hear about us?”
Woman: Yeah.
Man: But what are you hoping to get out of this? What are your goals even? How do you learn best?
Woman: I love that. It’s like actually caring if people get something out of it. Like, say it’s about the future of work or something. Wouldn’t you be way more into it if you had a say in what they even talked about?
Man: Totally. And picture this. What if there were, like, online forums, but before the conference?
Woman: Okay.
Man: People connecting, saying what they’re hoping for, starting the conversation early. By the time they get there, they already know people.
Woman: No more awkward icebreakers. You’re already halfway there.
Man: Exactly. And then the conference itself. Ditch those long, boring presentations, right? Shorter, punchier, more hands on stuff, workshops, solving problems together.
Woman: Instead of just sitting there, you’re doing something, working with people, trying out new ideas, so much better.
Man: And feedback. Not just those forms you forget on the way out.
Woman: Yeah.
Man: But all the way through, telling them what’s working, what’s not. They actually listen and make it better.
Woman: So you’re always tweaking, making it better as you go, based on what people actually need.
Man: Exactly. People feel like they’re part of it. Like their voice matters. That’s community, right there.
Woman: So, bottom line, maybe it’s time we stop just accepting the same old conference, you know?
Man: It’s time to expect more.
Woman: We can be those users demanding better, and who knows? Maybe change the whole game. That’s what I’m taking away from this.
Man: Me too. It’s about potential. Conferences could be so much more than we settle for.
Woman: Totally. Thanks for diving deep with us. Hope this got the wheels turning about what these things could be. See you next time.

How NotebookLM’s Audio Overview went wrong

Everything I wrote is wonderful!

My post about how the responsibilities of conferences mirror those of media platforms was an opinion piece. NotebookLM’s Audio Overview turned it into a PR flack’s dream.

Can NotebookLM be critical? Does it ever respond: “I can’t summarize this post intelligently or accurately?”

Somehow, I doubt it.

Despite sounding fluent and convincing, the generated “conversation” was riddled with inaccuracies. Here are some major issues:

Who wrote the blog post?

Hey, Audio Overview, I wrote the blog post, not Jeff Jarvis!

Misrepresentation of the responsibilities of platforms vs. conferences

The “woman” in the conversation says the blog post compares designing online platforms like Google and Twitter to designing conferences. This isn’t accurate. My post compares the responsibilities of media platforms and conferences, not their designs.

Exaggeration of user control

The conversation suggests that platforms like Google or Twitter are “all about putting you in control,” implying complete user freedom. However, my post and Jeff Jarvis’s quote are more nuanced. While Jarvis advocates giving users power and flexibility, there’s no suggestion that media platforms fully hand over control to users.

Invention of new topics: data portability, apps, and live feedback

The audio invents topics I never mentioned, such as data portability (taking contacts and files when switching platforms), a high-tech conference app where participants take notes and pictures, and vague musings about live feedback during the event. While these are interesting ideas, they have no basis in the original post, where I discuss the portability of the peer conference process, not technology.

Fabrication of pre-conference engagement

The “man” talks about sending surveys to conference attendees before the event, to ask what they want to learn. While this could be a useful idea, I didn’t mention anything about pre-conference surveys in the blog post. The blog discusses uncovering participants’ wants and needs during the event, not beforehand.

Conclusions

Audio Overview in Google’s NotebookLM generates bogus output. The tool transformed my blog post into an overly simplified, inaccurate PR pitch for conference engagement. What concerns me most is that it does this convincingly by aping the conversations humans have used to communicate and connect for over a hundred thousand years. Consequently, users may not question the validity of what is said because the format is familiar and comfortable.

But wait, there’s more! In my next post, I’ll explore how the Audio Overview service handles factually incorrect content. Stay tuned for my next test!

ChatGPT is just a tool

A photograph of a hammer labeled “ChatGPT”Sorry folks, but ChatGPT is just a tool, like a hammer, pen, spell checker, or Grammarly. Like any tool, ChatGPT can be useful—and it can also be dangerous when used incorrectly.

ChatGPT is a tool called a large language model (LLM). However, marketers love to call ChatGPT and other LLMs artificial intelligence (AI) because this framing ties into our culture’s fascination with understanding who we are, how we’re able to do what we do, and whether we are unique in our abilities or not. This fascination causes us to give human names to familiar objects, like cars, and to anthropomorphize our pets and tools. Harnessing fascination is a good way to sell us stuff.

Artificial intelligence?

I’ve programmed computers for over half a century and taught college computer science from 1983 – 93. Computer researchers began to use the term artificial intelligence in the 1940s and 50s.

In 1993, JET Thomas and I conducted a seminar at Marlboro College on AI. At that time, a second AI winter held sway. Rule and frame-based systems predominated. Researchers had not implemented key components of effective neural networks, our current core AI technology. The best chess-playing programs could beat most players, but Grandmaster Gary Kasparov wouldn’t lose a chess match to a machine until 1997. During the seminar, teachers and students alike shared a fascination with the concept of AI. However, the technology of the time generated output that was completely determined by the rules built into its programs.

However, unlike the software available during our 1993 seminar, humans can’t determine how today’s LLMs come up with their responses. And these sophisticated neural networks, often called deep learning, are everywhere. We use them for speech and image recognition, and natural language processing. ChatGPT has been used to “write and debug computer programs, compose music, teleplays, fairy tales and student essays, answer test questions (sometimes, depending on the test, at a level above the average human test-taker), generate business ideas, write poetry and song lyrics, translate and summarize text, emulate a Linux system, simulate entire chat rooms, play games like tic-tac-toe, or simulate an ATM.”

Just because ChatGPT can be asked to do these things doesn’t guarantee that it creates reliable answers. The software’s creator, OpenAI, acknowledges that ChatGPT “sometimes writes plausible-sounding but incorrect or nonsensical answers”.

Here’s a simple example of the limitations and unreliability of ChatGPT.

The limitations and unreliability of ChatGPT

ChatGPT is just a tool
I asked ChatGPT to rewrite a passive voice sentence in an active voice.


Me: rewrite “Try not to get stuck.” as an active sentence.

ChatGPT: ”Make an effort to avoid getting stuck.”

Me: That’s still a passive sentence!

ChatGPT: Apologies for that oversight. Here’s an active sentence version:

“Strive to avoid getting stuck.”

Me: Still passive!

ChatGPT: I apologize for the continued oversight. Here’s an active sentence version:

“Take proactive steps to avoid getting stuck.”

Me: “getting stuck” is passive, you’ve used it each time. rewrite it.

ChatGPT: Apologies once again for the oversight. Here’s an active sentence with a revised phrase:

“Make an effort to prevent becoming trapped.”

Me: “becoming trapped” is passive too. You’ve failed.

ChatGPT: Apologies for the repeated oversight. Here’s an active sentence with an active phrase:

“Exert effort to evade getting trapped.”

Me: Grammar is not your strong point.

ChatGPT: I apologize for the errors in my previous responses. I appreciate your patience, and I’ll make sure to provide a correct and active sentence for you. Here’s an active version of the sentence:

“Strive to avoid getting stuck.”


<Sigh>.

An automated mansplaining machine

This and numerous other examples of the confident fabrications of ChatGPT have led many to characterize it—fairly in my view—as “an automated mansplaining machine“, “autocomplete on steroids“, and a “bias optimizer“. As the third reference points out, “ChatGPT cannot recontextualize or independently seek out new information that contradicts [its] built-in assumptions.”

In the above example, it would have been refreshing if ChatGPT said at some point, “Sorry, I don’t know how to do that.” Instead, it repeatedly declared it had the answer. Hence the danger of using this tool. You have no guarantee anything ChatGPT avows is actually true! How can we use something we can’t completely trust?

The answer is: Carefully!

ChatGPT and the future of work

In 2019, I wrote about my concerns about the impact of machine learning on the future of work. Recently, people have written thousands of articles about how LLMs like ChatGPT jeopardize the future work prospects of numerous professions.

Some bosses will undoubtedly believe that they can replace some employees with fewer people writing well-prompted responses from ChatGPT. Such beliefs are dangerously naive. Here are three examples of the problems that can arise:

All new technology transforms job scope over time. But only some, like the automobile, eliminate industries. ChatGPT, by making some forms of work more efficient may allow employees to be more productive, meaning a static industry may be able to employ fewer workers. But I don’t see the wholesale future elimination of lawyers, writers, software makers, and other professions that many predict.

ChatGPT is just a tool

My advice is to think of ChatGPT as just a tool. Knowing when and how to use a specific tool is important. For example, as I write this post I’m using Grammarly to check my spelling and grammar. It’s a useful tool that catches most of my errors on the fly and occasionally improves how I express myself. However, it often makes suggestions that are simply incorrect and need to be suppressed. If I relied on Grammarly 100% to proofread my work, errors would slip through. (Regardless, I’m not claiming I’m a perfect editor of my own work!)

Similarly, I find ChatGPT useful for suggesting possible blog post titles, summarizing articles, and brainstorming ways to express ideas in print. However, most of the LLM’s suggestions are not helpful to me; I always need to review them to determine if there’s anything worth using.

In conclusion, treat ChatGPT as a useful but fallible tool, rather than an oracle. And don’t worry too much about its impact on your future work prospects!

AlphaZero, machine learning, and the future of work

Not long ago I wrote about the end of decent-paid jobs and the need for basic income. A startling recent advance in machine learning has only heightened my concerns about the future of work. Last month, Google’s subsidiary, DeepMind, published a paper on AlphaZero, an artificial intelligence (AI) the company designed to play games. The AI started with only game rules. Here’s what happened next:

“At first it made random moves. Then it started learning through self-play. Over the course of nine hours, the chess version of the program played forty-four million games against itself on a massive cluster of specialized Google hardware. After two hours, it began performing better than human players; after four, it was beating the best chess engine in the world.”
—James Somers, New Yorker, How the Artificial-Intelligence Program AlphaZero Mastered Its Games

From “knowing” nothing about the game, in four hours the program became the strongest chess player in the world. AlphaZero also taught itself to become the world’s best Go and shogi player in a few hours.

As a schoolboy, I played competitive chess for a few years. Although I haven’t played chess seriously since then, I still have a feeling for the game.

I was shocked watching AlphaZero’s tenth game with Stockfish, the strongest open-source chess engine in the world.

I’d describe AlphaZero’s play as completely solid, interspersed with incredible flashes of brilliance. Great human chess players have an uncanny ability to view a position and quickly select a few plausible moves for deeper study out of the many possible legal moves. The best grandmasters occasionally discover a brilliant and unexpected move in a game. AlphaZero found several during this game.

Having seen this game, I’d describe AlphaZero as the most creative, brilliant, and strongest chess player the world has ever seen.

From a novice to best in the world in four hours, is a level of performance that no human can match.

Now think about what would happen if this kind of performance could be achieved in human work environments such as:

  • medical scan diagnosis;
  • legal document creation;
  • engineering design; and
  • stock market trading.

These are only harder problems than playing a game because:

  • the problem space is larger; and
  • the data needed for learning can’t be self-generated by the AI itself and must be supplied by humans.

But these are not insuperable obstacles. If overcome, many high-paid jobs for medical practitioners, lawyers, accountants, and financial analysts would disappear.

So what is the future of work? Are we moving towards a world where the only available work is in low-paid “human service” areas where people are still cheaper than machines? Perhaps.

Until the arrival of robots capable of doing just about everything humans do. What work for humans remains then?

Impediments to AI matchmaking at events

artificial intelligence matchmaking at events: a screenshot from the movie "Star Wars" of the famous alien bar sceneCompanies are now marketing services for artificial intelligence matchmaking at events. However, unresolved issues could impede the adoption of this technology, especially by attendees.

Consider this marketing pitch for an artificial intelligence event matchmaking service:

“Using the [AI] platform…it’s easier for attendees to make sure they have the right meetings set up, and for exhibitors to have a higher return on investment in terms of connections with high-quality buyers.”
—Tim Groot, CEO Grip, as quoted in What AI Means To Meetings: How Artificial Intelligence will boost ROI, Michael Shapiro, July 2017 Meetings & Conventions Magazine

A win-win for exhibitors and attendees?

Tim describes using artificial intelligence matchmaking at events as a win for both exhibitors and attendees.

I’m skeptical.

Let’s assume, for the moment, that the technology actually works. If so, I think suppliers will reap most of the touted benefits, quite possibly at the expense of attendees. Here’s why.

Successful matchmaking needs digital data about attendees. An AI platform cannot work without this information. Where will the data come from? Tim explains that his service builds a profile for each attendee. Sources include “LinkedIn, Google, and Facebook”, while also “scouring the web for additional information”.

Using social media platform information, even if attendee approval is requested first, creates a slippery slope, as privacy issues in meeting apps remain largely undiscussed and little considered by attendees during the rush of registration. The end result is that the AI matchmaking platform gains a rich reservoir of data about attendees that, without strong verifiable safeguards, may be sold to third parties or even given to suppliers.

In addition, let’s assume that exhibitors get great information about whom to target. The result: “high-value” attendees will be bombarded with even more meeting requests while attendees who don’t fit the platform’s predictions will be neglected.

In my opinion, the best and most likely to succeed third-party services for meetings are those that provide win-win outcomes for everyone concerned. Unfortunately, it’s common (and often self-serving) to overlook a core question about meeting objectives —whom is your event for? — and end up with a “solution” that benefits one set of stakeholders over another.

How well will artificial intelligence matchmaking at events work for attendees?

Artificial intelligence is hot these days, so it’s inevitable that event companies talk about incorporating it into their products, if only because it’s a surefire way to get attention from the meetings industry.

I know something about AI because in the ’80s I was a professor of computer science, and the theory of artificial neural networks — the heart of modern machine learning — was thirty years old. AI had to wait, however, for the introduction of vastly more potent technology to allow practical implementation on today’s computers.

While the combination of powerful computing and well-established AI research is demonstrating incredible progress in areas such as real-time natural language processing and translation, I don’t see why sucking social media and registration data into a database and using AI to look for correlations is going to provide attendee matchmaking that is superior to what can be achieved using participant-driven and participation-rich meeting process combined with attendees’ real-time event experience. (Once again, exhibitors may see a benefit from customized target attendee lists, but I’m looking for a win-win here.)

From the attendee’s point of view

When attendees enter a meeting room there’s a wealth of information available to help make relevant connections. Friends introduce me to people I haven’t yet met. Eavesdropping on conversations opens up more possibilities. Body language and social groupings also provide important potential matchmaking information. An AI matchmaking database includes none of these resources. All of them have led me (and just about everyone who’s ever attended meetings) to professional connections that matter.

Coda

I’ll conclude with a story. The June 2017 PCMA Convene article Can Artificial Intelligence Make You a Better Networker? describes a techsytalk session by Howard Givner where he “gave particular emphasis to the importance of facilitated matchmaking at events.” I like to think that Howard discovered this when he attended the participant-driven and participation-rich EventCamp East Coast I designed and facilitated in 2010, about which he wrote:

“…it was one of the most innovative and eye-opening professional experiences I’ve had. Aside from coming back with lots of new tips and ideas, I easily established triple the number of new contacts, and formed stronger relationships with them, than at any other conference I’ve been to.”

We didn’t use an AI matchmaking service.