Worldcoin deep dive: saving us from toxic AI sludge?

Back in  2017, among the endless carrousel of new, promising projects that rarely amounted to anything, was one called Civic.

Civic was not a solution in search of a problem, which is what most of the projects back then amounted to.

Civic actually had a great solution to a serious problem.

It wanted to use blockchain to secure a person’s identity, so they could do away with passwords and logins.

Kind of like how you can use your Google account to log into things now, but the idea was you could use it for everything. 

It would also have the advantage of not being controlled by Google, or anyone else – at least in theory.

Remember, Google has a long, long history of ending services and leaving people in the lurch – think Google Domains, Google Podcasts, Google Play Music, Google Stadia, Google Hangouts, G Suite, Nest, Google+, Google Spaces, Picasa, Google Talk… the list is endless.

Google has killed 295 services so far.

So I wouldn’t trust Gmail or Google login to last forever. Google is not a good company to rely on…

But I guess that’s beside the point. 

Anyway, Civic is still going, but it’s also kind of dead. I just had a look on its website and it still doesn’t seem to have released its flagship product, Civic ID.

Civic has been around since 2015, and its token has been around since 2017. So it’s had at least seven years to get it out into the world.

 Seven years! 

That’s longer than it took to build the Hoover Dam or the Golden Gate Bridge.

It’s longer than the entire second world war.

It’s almost as long as IOTA has been promising coordicide (lol).

Anyway…

With the advent of generative AI – think ChatGPT – the need for people to have an immutable way to prove their identity has resurfaced.

It goes back to what Near’s co-founder, Illia Polosukhin, said in an interview with CoinDesk last year (as I reported in my Near deep dive):

I think AI right now will become a tool to break a lot of the current societal systems. So you can [overwhelm] the legal system with frivolous lawsuits. You can create fake candidates and try to get them elected. You can create fake companies and raise money for them.

I do find it interesting that people’s main fears about AI seem to be that it will either take all our jobs (which does feel likely) or evolve into a superintelligence and kill us all (a la Terminator). 

But I haven’t heard many people talking about what Polosukhin is saying here. And that is much easier to do and much more likely to happen. It’s kind of inevitable if you think about it.

The internet is already getting overwhelmed with fake content. And it’s been happening since long before AI made it possible to crank out a 3,000-word Search Engine Optimised article about the best toaster to buy.

That started with something called affiliate marketing. Basically companies pay websites a commission if someone comes from a website and then makes a purchase.

So if you can make your website the top hit for Google when people search “what’s the best toaster”, and then you link to a store selling it, you will get a cut of the sale.

And these commissions can really add up.

The more popular the search term and the higher your site ranks, the more you will make.

And the way you get your website ranked higher in Google is to write 3,000 word articles that hit on all the key search terms.

(Kind of ironic that I’m writing so much in my intro here, isn’t it? In my defence, these premium articles are paywalled, so they don’t even really get indexed by Google.)

People really started getting into this about 10 years ago.

Which is why now, if you search for anything, you can’t find the information you need without sifting through a 3,000-word article for the one sentence you need.

The other 2,980 words are pure bloat.

(The solution to this is just to put Reddit after anything you’re searching for and you’ll usually get to a helpful reddit thread on it where the best answer is upvoted to the top. But the bots are coming for Reddit, too. So this won’t last forever.)

It used to be that people would use sites like upwork and fiverr to pay someone living in a poor country to write these articles for a pittance.

But even that cost money and took time.

Now, you can use AI to crank out an article in seconds.

And so the internet will soon – likely already is – become almost completely useless. At least in terms of getting information.

Plus, you can use AI to make bots that can (and are) flooding platforms like Reddit, Twitter, Instagram, etc.

Something called Dead Internet Theory started as a kind of jokey conspiracy a few years ago, but with the advent of generative AI, it’s now essentially coming true.

You can read about how that happened here.

But this is about even more than just the internet.

As Polosukhin said, you can now use AI to do to our real world systems what affiliate marketing did to the internet.

So, what is the solution?

Blockchain of course!

Again, from Polosukhin:

The reality is most of the solutions will use Web3 because you need authentication, you need identity, you need traceability, you need all those pieces embedded across the whole spectrum of our society. …

Even like the articles you guys [CoinDesk] writes, you need to have cryptographic signatures on the articles to understand that they have not been modified by somebody else. Or like me cosigning it, that you wrote the article and I agree with it. Stuff like that.

One man who knows this all too well is the man who unleashed ChatGPT upon the world – Sam Altman.

And he has created a crypto project – Worldcoin – to (try) solve that problem.

So let’s take a look at it…