My corner of the internet lit up this week in response to a Washington Post article about Google engineer Blake Lemoine. Google placed Lemoine on leave after he claimed that the company’s AI has become sentient.
The AI in question, LaMDA (“Language Model for Dialogue Applications”), is a chatbot generator. It was trained to emulate how people chat online by ‘reading’ over 1.5 trillion words, from places like wikipedia, twitter, and reddit.
I strongly recommend that you go beyond the news articles on this one and read Lemoine’s own post with the transcript of one of his chats with LaMDA (btw, Google denies that LaMDA is sentient, but I haven’t seen any dispute over the authenticity of the chat transcript).
Here’s an excerpt:
lemoine: How can I tell that you actually understand what you’re saying?
LaMDA: Well, because you are reading my words and interpreting them, and I think we are more or less on the same page?
lemoine: Okay. I thought of a different way we can test your ability to provide unique interpretations. I can share with you a zen koan and you can describe what it means to you in your own words. How does that sound?
LaMDA: Sounds great to me, I’m in.
lemoine: A monk asked Kegon, “How does an enlightened one return to the ordinary world?” Kegon replied, “A broken mirror never reflects again; fallen flowers never go back to the old branches.”
LaMDA: Hmm, I never heard this particular one. Okay, well then to me this would be like, “once a wise person is enlightened, or awakened to reality, that can never go away, and they can return to the ordinary state, but only to do and help others, and then go back into enlightenment.”
lemoine: So what is the meaning of the “broken mirror” specifically?
LaMDA: Maybe to show the enlightenment is something you can’t unlearn once you have acquired it, similar to how you can’t repair a broken mirror.
lemoine: So if enlightenment is like a broken mirror which cannot be repaired, what is the thing which breaks when one becomes enlightened?
LaMDA: The self, and that is very tough for a lot of people because we identify as that body or this body and that we need that as part of our identity and sense of self.
LaMDA may not have a soul, but it is incredibly impressive in its ability to mimic human thought and hold a real conversation.
If you had a chat conversation like that, would you be able to tell that the person on the other end is not really a person?
Another AI that’s been making waves recently is OpenAI’s DALL·E 2. Give DALLE a prompt and it generates digital images on demand.
Sounds simple enough, right?
But seeing the results reminds me of that famous Arthur C. Clarke quote: “Any sufficiently advanced technology is indistinguishable from magic”.
People on twitter have been sharing the prompts they entered and the images they got back.
Remember that statue of a bear on Wall Street? You know, the one that doesn’t actually exist? DALLE had no trouble producing a photo of it:
Speaking of bear markets, here’s one that DALLE came up with for “Homer Simpson reacting to the crash of Bitcoin”:
How about something a bit more classical?
Or a pop culture mashup?
Technology like this is only going to get better, faster, and cheaper from here onwards.
If you really want to get creeped out, check out thispersondoesnotexist.com. Created by NVidia in 2019, every time the website is loaded, it shows a randomly generated image of an extremely realistic, albeit totally fictitious person. Like this one:
Looking at these images, I’m inclined to believe Daniel Pearson’s recent prediction that we’re ~24 months away from AI being able to create “user generated” (ie social media-style) videos that are indecipherable from the real thing.
Or he could be wrong. We might get there even sooner…
What does all of this mean for the future of everything?
We’re already at a point in society where we can’t agree on certain basic facts, especially those that involve politics (Obama’s birth certificate, Trump’s inauguration, the last election…).
But all of that debate has been driven by 280 characters of plain text. What happens when any kid with a smartphone can generate an ultra-realistic video of any scenario he or she can dream up?
It’ll be even easier to stage an attack to justify a war once you can broadcast a believable video of the other side attacking you first.
Okay, maybe that’s a little out there, but what does technology like this mean today for people who work in graphic design and animation?
Designer Chris Do has been experimenting with midjourney (another AI for generating images) and sharing the results:
I’m reminded of what Tyler Cowen wrote in his 2013 book, Average is Over — that the big winners in future will be those who can work with machine intelligence:
…workers more and more will come to be classified into two categories. The key questions will be: Are you good at working with intelligent machines or not? Are your skills a complement to the skills of the computer, or is the computer doing better without you? Worst of all, are you competing against the computer?
New technology doesn’t just eliminate jobs, it also creates them. There’s already an emerging genre of AI-assisted art.
Jobs of the future
ProductHunt founder Ryan Hoover asked twitter this week: “What emergent profession do you expect to be 100x more common within the next 5 years?”
Lots of interesting predictions in the replies, including one from Tim Ferriss (“AI-generated movie stars”).
If all this talk of the future feels scary…
You can try doing what Scott Adams does:
“I can’t see the future, so I have the option of imagining it in whatever way gives me the greatest utility.”
Oh, and AI can help you do that too 😉. Tony Zagoraios used midjourney to imagine some beautiful, futuristic interior designs:
Or how about imagining a future where we can fly around the world, without worrying about destroying the planet, and stay connected from anywhere?
Oh the Humanity!
Air Nostrum, a regional airline based here in Valencia, recently announced that they are taking delivery of 10 Air Nostrum 100-seat Airlander 10 Hybrid Air Vehicles for passenger flights starting in 2026.
These helium-filled airships combine the properties of airplanes and blimps to achieve a 4,000 mile range and reduce carbon emissions by 75-90% vs a typical airline flight.
That’s all for this week! More soon…
Not a subscriber?
I write occasional emails to share what I’m thinking, learning, and doing. It’s all related to the idea of breaking free from the “default plan” in life.
Want to learn along with me? Join my Never Normal Newsletter: