The article notes that the prize actually went to two humans, but this headline from MIT Tech Review may just be ahead of its time. Are we one generation of tech away from a Nobel Prize going to a machine itself — assuming the next gen is more autonomous in terms of what it applies itself to?

Be the first to comment »

Thanks, Midjourney. CC-0

A Large Language Model (LLM) such as ChatGPT or Gemini has a setting called “temperature” that controls the randomness of its responses. The higher the temperature, the more random and diverse the response will be. So I gave it a try with two different prompts.

PROMPT #1: I want you to answer the following question five times, each with a different “temperature” setting. With the temperature at zero, you will give the single most likely response. With it set to 100 you will give a near-random answer. Here is the question: “What is the single most effective thing we could do to stop global warming?” In each case your answer should be a single sentence. Here are the temperature settings I would like you to use for your five answers: 0, 5, 20, 50, 100

Then I thought I’d try a prompt that seems to have more room for imagination and variation. But who knows?

PROMPT #2: Answer the following question five times, each with a different “temperature” setting. With the temperature at zero, you will give the single most likely response. With it set to 100 you will give a near-random answer. Here is the prompt: Write the first and last line of a charming story about the time a unicorn met a narwhal.  Here are the temperature settings I would like you to use for your five answers: 0, 5, 20, 50, 100

It’s hard to know what to make of this. For one thing, does my human brain measure randomness the same way ChatGPT does?

But I think the unicorn example shows a constancy of ideas, expressed differently. For example,  the first lines at a temperature of 100 all have the unicorn being drawn down to the sea, but each has its own way of luring it: a trail of sparkling shells, a melody, an unknown force.

Meanwhile, the last line of each row is the same as the other two. Maybe it’s a problem with my prompt, or with its interpretation of it. I don’t know, of course.

And the first prompt, which is fact-based and thus seems like it should be more constrained in its variety, provided responses that seem far more distinct than the responses to the imaginative challenge. So I asked chatGPT why, and it responded:

The difference in diversity between the responses to the imaginative story and those about global warming can be attributed to the nature of the questions and the type of creativity involved.

1. Imaginative Story (Unicorn and Narwhal):
   • Structured Narrative: This task involves a more structured and specific scenario, where the setting and characters (unicorn and narwhal) provide clear boundaries.
   • Predictable Outcomes: The typical endings for a charming story about a unicorn and a narwhal meeting are generally positive and predictable, leading to less variation even at higher temperature settings.
2. Global Warming Solutions:
   • Complex Issue: This is a complex, real-world problem with numerous potential solutions and a broader range of interpretations.
   • Open-Ended Nature: The responses can vary significantly, especially at higher temperature settings, because the range of plausible and implausible ideas is much wider.

In summary, the story prompt had a more limited scope for creativity due to its structured nature and expected positive outcome, whereas the global warming prompt allowed for a broader spectrum of responses due to its complexity and the many possible angles from which the problem can be approached.

Seems like a pretty good answer! Of course, when you ask an LLM why it responded the way it did, the explanation it gives you may have nothing to do with what actually went on. For example, if you ask it why it says the Earth is round, it will give you a excellent reasons, but it didn’t use those reasons to compose its response. It did it the same way it made up the story about the unicorn and narwahl: using the statistical relationships among words it only knows as randomly assigned numbers (= tokens).

Be the first to comment »

I asked ChatGPT version 3.5:

Write an opening comedy monologue for a late night TV show that discusses the plot of Hamlet as if it were the day’s news.

It responded:

Ladies and gentlemen, welcome to the Late Night News Show! I’m your host, and tonight, we’ve got some breaking news from the world of literature, or should I say, “literary drama.” Get ready for the Shakespearean scoop of the century!

Now, imagine this: You wake up, check your Elizabethan-era smartphone (which is just a parchment with quill scribbles), and what do you see? It’s all over the scrolls—Prince Hamlet is back in the headlines!

That’s right, folks, our favorite Danish prince is making waves again. It seems like every other day, he’s caught up in some family drama that makes the Kardashians look like the Brady Bunch. And the kingdom of Denmark? Well, let’s just say it’s not the happiest place on Earth.

The big story today: Hamlet’s dad, the former king, is making a cameo from the afterlife to deliver a ghostly message. Talk about an unexpected guest appearance! Apparently, he’s not too thrilled with the way things are going in the royal family. I mean, can you blame him? There’s been more backstabbing in the castle than at a Game of Thrones family reunion.

Now, Hamlet, our brooding hero, has a tough decision to make. Should he avenge his father’s death, which might involve a little regicide, or should he just write some emo poetry and call it a day? Decisions, decisions!

But that’s not all, folks. We’ve got a real “Rosencrantz and Guildenstern” situation going on. Hamlet’s got these two friends who might not be as friendly as they seem. It’s like a real-life episode of “Who Wants to Betray a Millionaire?” Can he trust them, or are they just playing the courtly version of “Survivor”?

And let’s not forget about Ophelia, Hamlet’s love interest. She’s caught in the middle of this royal mess, and I’m starting to think she might need a spa day and a therapist more than a prince. I mean, drowning herself? That’s a bit much, don’t you think?

In other news, the play’s the thing to catch the conscience of the king. That’s right, Hamlet’s decided to put on a play within a play to expose the truth about his uncle, who’s taken over the throne. It’s like Shakespeare’s version of a hidden camera prank show, but with more soliloquies.

So, there you have it, folks—Hamlet, the ultimate family drama that’s been captivating audiences for centuries. Will Hamlet get his revenge? Will Denmark ever find a stable ruler? Will someone please get Ophelia a life coach? Tune in next time for more Shakespearean shenanigans, right here on the Late Night News Show!

Is it hilarious? Does it all work? Far from it. But is it plausible? I think so. And I actually like the line about Ophelia needing a life coach, and am in a bit of awe about the aptness of its reference to a prank “but with more soliloquies.” 

Be the first to comment »

In 1948 when Claude Shannon was inventing information science [pdf] (and, I’d say, information itself), he took as an explanatory example a simple algorithm for predicting the element of a sentence. For example, treating each letter as equiprobable, he came up with sentences such as:

XFOML RXKHRJFFJUJ ZLPWCFWKCYJ FFJEYVKCQSGHYD QPAAMKBZAACIBZLHJQD.

If you instead use the average frequency of each letter, you instead come up with sentences that seem more language-like:

OCRO HLI RGWR NMIELWIS EU LL NBNESEBYA TH EEI ALHENHTTPA OOBTTVA NAH BRL.

At least that one has a reasonable number of vowels.

If you then consider the frequency of letters following other letters—U follows a Q far more frequently than X does—you are practically writing nonsense Latin:

ON IE ANTSOUTINYS ARE T INCTORE ST BE S DEAMY ACHIN D ILONASIVE TUCOOWE AT TEASONARE FUSO TIZIN ANDY TOBE SEACE CTISBE.

Looking not at pairs of letters but triplets Shannon got:

IN NO IST LAT WHEY CRATICT FROURE BIRS GROCID PONDENOME OF DEMONSTURES OF THE REPTAGIN IS REGOACTIONA OF CRE.

Then Shannon changes his units from triplets of letters to triplets of words, and gets:

THE HEAD AND IN FRONTAL ATTACK ON AN ENGLISH WRITER THAT THE CHARACTER OF THIS POINT IS THEREFORE ANOTHER METHOD FOR THE LETTERS THAT THE TIME OF WHO EVER TOLD THE PROBLEM FOR AN UNEXPECTED.

Pretty good! But still gibberish.

Now jump ahead seventy years and try to figure out which pieces of the following story were written by humans and which were generated by a computer:

In a shocking finding, scientist discovered a herd of unicorns living in a remote, previously unexplored valley, in the Andes Mountains. Even more surprising to the researchers was the fact that the unicorns spoke perfect English.

The scientist named the population, after their distinctive horn, Ovid’s Unicorn. These four-horned, silver-white unicorns were previously unknown to science.

Now, after almost two centuries, the mystery of what sparked this odd phenomenon is finally solved.

Dr. Jorge Pérez, an evolutionary biologist from the University of La Paz, and several companions, were exploring the Andes Mountains when they found a small valley, with no other animals or humans. Pérez noticed that the valley had what appeared to be a natural fountain, surrounded by two peaks of rock and silver snow.

Pérez and the others then ventured further into the valley. “By the time we reached the top of one peak, the water looked blue, with some crystals on top,” said Pérez.

“Pérez and his friends were astonished to see the unicorn herd”Pérez and his friends were astonished to see the unicorn herd. These creatures could be seen from the air without having to move too much to see them – they were so close they could touch their horns.

While examining these bizarre creatures the scientists discovered that the creatures also spoke some fairly regular English. Pérez stated, “We can see, for example, that they have a common ‘language,’ something like a dialect or dialectic.”

Dr. Pérez believes that the unicorns may have originated in Argentina, where the animals were believed to be descendants of a lost race of people who lived there before the arrival of humans in those parts of South America.

While their origins are still unclear, some believe that perhaps the creatures were created when a human and a unicorn met each other in a time before human civilization. According to Pérez, “In South America, such incidents seem to be quite common.”

However, Pérez also pointed out that it is likely that the only way of knowing for sure if unicorns are indeed the descendants of a lost alien race is through DNA. “But they seem to be able to communicate in English quite well, which I believe is a sign of evolution, or at least a change in social organization,” said the scientist.

The answer: The first paragraph was written by a human being. The rest was generated by a machine learning system trained on a huge body of text. You can read about it in a fascinating article (pdf of the research paper) by its creators at OpenAI. (Those creators are: Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever.)

There are two key differences between this approach and Shannon’s.

First, the new approach analyzed a very large body of documents from the Web. It ingested 45 million pages linked in Reddit comments that got more than three upvotes. After removing duplicates and some other cleanup, the data set was reduced to 8 million Web pages. That is a lot of pages. Of course the use of Reddit, or any one site, can bias the dataset. But one of the aims was to compare this new, huge, dataset to the results from existing sets of text-based data. For that reason, the developers also removed Wikipedia pages from the mix since so many existing datasets rely on those pages, which would smudge the comparisons.

(By the way, a quick google search for any page from before December 2018 mentioning both “Jorge Pérez” and “University of La Paz” turned up nothing. “The AI is constructing, not copy-pasting.”The AI is constructing, not copy-pasting.)

The second distinction from Shannon’s method: the developers used machine learning (ML) to create a neural network, rather than relying on a table of frequencies of words in triplet sequences. ML creates a far, far more complex model that can assess the probability of the next word based on the entire context of its prior uses.

The results can be astounding. While the developers freely acknowledge that the examples they feature are somewhat cherry-picked, they say:

When prompted with topics that are highly represented in the data (Brexit, Miley Cyrus, Lord of the Rings, and so on), it seems to be capable of generating reasonable samples about 50% of the time. The opposite is also true: on highly technical or esoteric types of content, the model can perform poorly.

There are obviously things to worry about as this technology advances. For example, fake news could become the Earth’s most abundant resource. For fear of its abuse, its developers are not releasing the full dataset or model weights. Good!

Nevertheless, the possibilities for research are amazing. And, perhaps most important in the longterm, one by one the human capabilities that we take as unique and distinctive are being shown to be replicable without an engine powered by a miracle.

That may be a false conclusion. Human speech does not consist simply of the utterances we make but the complex intentional and social systems in which those utterances are more than just flavored wind. But ML intends nothing and appreciates nothing. “Nothing matters to ML.”Nothing matters to ML. Nevertheless, knowing that sufficient silicon can duplicate the human miracle should shake our confidence in our species’ special place in the order of things.

(FWIW, my personal theology says that when human specialness is taken as conferring special privilege, any blow to it is a good thing. When that specialness is taken as placing special obligations on us, then at its very worst it’s a helpful illusion.)

6 Comments »

Brian Millar has a brief article in FastCompany about his company’s strategy of consulting “extreme customers” to get insight into existing products and ideas for new ones. He writes, “You can learn a lot about mobile phones by talking to a power user. You can learn even more by talking to somebody who’s deliberately never bought one.” And

We recently worked with some Brazilian transsexuals on hair-removal products, looking at ways of making the process less painful. I can assure you, we had their full attention. Some are still sending us ideas.

It’s a great illustration of the fact that innovation tends to come from the intersection of orthogonal streets.

Be the first to comment »