Categories
Uncategorized

A conversation with ChatGPT

This week OpenAI introduced a new conversational text generator called ChatGPT. Here is a direct transcript of a conversation I had with ChatGPT where I asked it to explain itself.

Explain ChatGPT in a single sentence

ChatGPT is a large language model that was trained using a combination of text-based conversations and task-oriented dialogue data to perform natural language processing tasks and generate human-like responses in conversation.

What is a large language model?

A large language model is a type of artificial intelligence that is trained on a large amount of text data to generate human-like text and to perform natural language processing tasks. These models are typically based on deep learning algorithms and are able to generate text that is similar to human writing or speech. They can be used for a wide range of applications, including generating text, summarizing documents, translating languages, and answering questions. Some examples of large language models include GPT-3, BERT, and ChatGPT.

Explain it as a limerick.

A large language model’s grand plan
Is to understand the ways of man
It processes vast amounts
Of text to make new accounts
Of human-like speech, if it can.

Will large language models replace indexed search?

It is unlikely that large language models will completely replace indexed search, as indexed search and large language models serve different purposes. Indexed search uses a pre-determined set of keywords and indexing rules to quickly find relevant information based on a specific query. Large language models, on the other hand, are able to generate human-like text and perform natural language processing tasks, but they do not necessarily rely on pre-determined keywords or indexing rules. Instead, they use the vast amount of text data they have been trained on to generate responses that are relevant to a given query. While large language models may be able to assist with search in some cases, they are not designed to replace indexed search entirely.

How can we be sure of this?

It is difficult to say with certainty that large language models will not completely replace indexed search, as the field of artificial intelligence and natural language processing is constantly evolving. However, based on the current capabilities and limitations of large language models, it is unlikely that they will be able to completely replace indexed search. Indexed search is designed specifically for quickly finding information based on pre-determined keywords and indexing rules, while large language models are trained on large amounts of text data to generate human-like text and perform natural language processing tasks. These are different purposes, and it is unlikely that large language models will be able to fully replicate the capabilities of indexed search. Additionally, indexed search has been widely used and refined for many years, so it is likely to continue to play a significant role in information retrieval.

Explain that again in two sentences

It is unlikely that large language models will completely replace indexed search because they serve different purposes. Indexed search is designed for quickly finding information based on pre-determined keywords and indexing rules, while large language models are trained on large amounts of text data to generate human-like text and perform natural language processing tasks.

Rewrite that as a Shakespeare sonnet

In this age of artificial might,
We see machines that can generate text,
But indexed search still holds its place,
For quickly finding information with grace.

Though large language models may assist,
In tasks that require language and list,
Their capabilities do differ,
From the indexed search’s specific method.

And so we can be sure, in this time,
That indexed search will not be left behind,
For though large language models may evolve,
Indexed search will still be part of the resolve.


It’s fun to play with and can feel like magic, but its risk lies in that it can give answers which seem authoritative but are wholly incorrect — although, to be fair, humans are more than capable of this too.

Categories
Uncategorized

Neural Radiance Fields forever

A few weeks ago, Google introduced Immersive View for Maps, a high-quality 3D flythrough of popular spots, which shows day / night and real-time weather conditions.

It’s pretty cool. But even cooler is the technology behind it.

NeRF (Neural Radiance Fields) is a new approach to photogrammetry.

Categories
Uncategorized

Gucci goes digital

Avatars, identity, digital fashion, and generational change

Gucci just launched its first digital-only sneaker. The Virtual 25 cost £11.99 through the Gucci app, and you can wear them in your social media photos, and on your VR Chat and Roblox avatars.

Lots of brands have been experimenting with digital fashion and cosmetics, like L’Oreal’s Signature Faces, but to date it’s mostly been stunts and experimentation. Now they’re taking their first steps into earning money from them; you can buy Puma in Zepeto, Adidas in Aglet, and Oscar de la Renta in Drest.

Categories
Uncategorized

Art on the Blockchain

This is me thinking out loud—I mean that almost literally, I recorded an earlier version of it one evening as voice memo. I’m not firm in my convictions on this piece and am very open to having my mind changed.


There’s been a lot of news recently about digital art sales using a technology called NFTs*. These are, to give a very, very layman’s explanation, digital tokens which prove the ownership (or transaction history) of a digital object. Digital art can be copied infinitely because it’s digital, but an NFT uses a blockchain to provide proof of provenance so no matter how many copies are in existence you can have the ‘one true original’, certified and validated, which gives it (artificial) scarcity and (theoretical) resale value.

It’s kind of like if a conceptual artist had made a physical work of art using a Mars bar wrapper; you could say, well, anybody can have a Mars bar wrapper, in fact I’ve got one in my pocket, so that’s worth whatever the art is worth. But value is what we assign to things. So there’s a collective agreement that the artist’s work is art (transcending the object itself) and the art world uses certificates of authenticity and ownership to prove it’s the ‘original’ and not just a Mars bar wrapper you’ve just bought from the shop. That certificate maintains that Mars bar wrapper’s value. And that’s what NFTs do for digital art.

I’m torn on this because, in theory, I think it sounds like a good idea—artists getting justly rewarded for their work! But two things give me pause: the environmental impact, and cryptocurrencies.

The issue of the environmental impact is raised because of the way the blockchain (Ethereum in this case) works, which is by using a lot of computational power which is incredibly—I mean, incrediblyenergy intensive.

Because [cryptocurrency] coins ask the investors of tomorrow to buy in at ever increasing computational power, we have ended up in a horrific spiralling excess of energy usage and ecological devastation.

Everest Pipkin

A lot of people claim this is going to be fixed—there are many good arguments in favour of this not being true, but there are also some far less resource-intensive alternative cryptocurrencies.

The other thing is that I don’t have faith that cryptocurrencies aren’t just a big pyramid scheme, and the people spending big money on digital art with NFTs tend to be people who invested in crypto early, when their currency wasn’t as valuable as it is now:

What you have right now are these crypto-millionaires who are trying to create assets with the money that they’ve accumulated. One of the only ways they can do that right now is through the NFT market, because it accepts cryptocurrency

Lucien Smith

These people stand to gain from more people buying the currency and driving its value up, because cryptocurrencies are essentially useless as actual currencies, only as vehicles for speculation. Or, more succinctly put:

NFTs use “art” to make crypto speculation seem pro-social.

Nathan Jurgenson

But putting aside the environmental concerns (as if that’s even an option!) and my own cryptocurrency skepticism, there’s a further critical question to be answered: do NFTs actually work? As in, will they be accepted as actually providing sufficient provenance for a thing to retain a value?

There’s no question that people will buy digital art which has an NFT because they’re doing that right now. But it’s the sell-on which will be the test of whether or not this system works; whether an NFT is sufficient for a secondary buyer to say, I accept that this token that you have is proof that this is a certified digital artwork from the artist and not a copy.

It requires collective belief for this to work. If the tokens are accepted as a genuine guarantee of authenticity by a secondary buyer, then it works. If it doesn’t, if the NFTs are simply not accepted as provenance after the initial sale, then the whole thing falls apart.

So whether or not people are paying hundreds of thousands pounds for pieces of digital art from the artist (or a broker) isn’t the test of whether or not this system works. The test is: does the art retain value (or appreciate in value)? Will secondary buyers and traders accept NFTs as proof? Will there be a collective belief in the system that sustains the market?

And further still: should the market be sustained at all? Is the big money pouring into NFTs, “a devastatingly lopsided, dangerous and unregulated market that exists solely as a form of exploitative mass delusion” as Luke Plunkett puts it, helping to prop up an art establishment that elevates certain art into high art just to maintain its own business?

Like I said before, the idea of artists getting fair reward for their art is a great idea. I’m just extremely wary of the crypto world being seen as the solution for that.


* NFT = non-fungible token; that is, a token which signifies a unique object.