Features

An analysis of crypto heists linked to North Korea reveals private key exploits yield the best results for the regime.

More than 70% of the crypto lost to North Korea-linked hacks since 2020 was stolen via private key exploits, according to Magazines analysis of data from the United Nations Security Council (UNSC) and DeFiLlama.

The combined figures suggest North Korea was responsible for about $2.4 billion of crypto heists since 2020, of which $1.69 billion was stolen due to compromised private keys.

These cybercrimes are often attributed to the Lazarus Group a notorious hacking syndicate allegedly backed by the North Korean state and allegedly support the hermit kingdoms weapons of mass destruction program.

Read more

Hip hop artists can make more money from an NFT than a year of Spotify streams. But there are plenty of other reasons rappers love crypto.

Fifty years since Kool Hercs legendary DJ set at his sisters party in the Bronx kickstarted hip hop, the world has changed dramatically.

Back in the day, labels and radio were the gatekeepers for which artists would break today, anyone with talent and a cell phone can record music, grow a fanbase, and build a sustainable career from their bedroom.

With all the attention crypto and NFTs received during the 2017 and 2021 bull runs, its no surprise that numerous rappers have dipped their toes in the space, with some seeing spectacular success: Snoop Dogg, Logic, Nas, Ghostface Killah and Eminem, among them. 

Read more

Anti-crypto politicians in Washington are weaponizing disputed claims about crypto funding Hamas to try and pass draconian new legislation.

The Israel-Gaza war has once again thrown the spotlight on crypto, with anti-crypto politicians seizing on exaggerated reports of crypto being used to finance terrorism to introduce harsh new legislation with the potential to crush the industry.

Three days after Hamas carried out its brutal Oct. 7 attack, The Wall Street Journal published an inflammatory article stating that in the past three years, U.S.-designated terrorist organizations such as Hamas, Palestinian Islamic Jihad and Hezbollah had raised $134 million in crypto.

The article later corrected following an online backlash became ammunition for the anti-crypto army in Washington, which cited it to push for ever greater restrictions on crypto.

That came to a head over the past week with a bipartisan bill called the Terrorism Financing Prevention Act, introduced on Dec. 8. It obliges the Treasury to identify foreign financial institutions and crypto platforms that have knowingly conducted transactions with U.S.-designated terrorist outfits and enables it to impose sanctions to restrict U.S. bank accounts and block transactions.

Senator Mitt Romney tied the bill specifically to the Israel-Gaza war:

Read more

Games overrun with bots just show bot owners care, claims Pixels founder. Plus we review Galaxy Fight Club, chat to Illuvium’s CEO and more.

Some think that bots in games is a sign of the apocalypse, or perhaps just the makers trying to fill up an empty venue to make it look popular.

But Pixels founder and CEO Luke Barwikowski says that conversely, if people arent trying to fill your game with bots, then it’s probably because the game isn’t exactly the talk of the town.

If people aren’t trying to bot your game it’s not because they can’t it’s because they don’t care enough to do it.

Read more

If the Digital Asset Anti-Money Laundering Act were to become law, many cryptocurrency providers would have to learn how to comply with the same regulations as traditional financial institutions.

Real bipartisan legislative efforts are rare in Washington, DC, these days, but Democratic Senators Elizabeth Warren and Joe Manchin and Republican Senators Lindsey Graham and Roger Marshall have managed to come together to co-sponsor a bill focused on crypto crime. 

According to the senators, the Digital Asset Anti-Money Laundering Act of 2023 aims to close loopholes in the nations Anti-Money Laundering rules. The bill would amend the Bank Secrecy Act and would designate a diverse range of digital asset providers as financial institutions. 

The Bank Secrecy Act establishes program, recordkeeping and reporting requirements for national banks, federal savings associations, federal branches and agencies of foreign banks. Digital asset providers would be required to adhere to many of the same regulations as traditional banks.

Read more

Increasing numbers of blockchain marketplaces offer crowdsourced data and compute for AI models — and can AI analysis improve fund returns?

Were rolling out genuine use cases for AI and crypto each day this week including reasons why you shouldnt necessarily believe the hype. Today get two for the price of one: Blockchain based AI marketplaces, and financial analysis.

It may not seem like the most exciting use case blending AI and crypto, but both Near co-founder Illia Polosukhin and Framework Ventures founder Vance Spencer cite blockchain-based marketplaces that source data and compute for AI as their top pick.

AI is an incredibly fast-growing industry requiring ever-increasing amounts of computing power. Microsoft alone is reportedly investing $50 billion into data center infrastructure in 2024 just to handle demand. AI also needs enormous amounts of raw data and training data, labeled into categories by humans.

Polosukhin believes decentralized blockchain-based marketplaces are the ideal solution to help crowdsource the required hardware and data. 

You can use [blockchain] to build more effective marketplaces that are more equal, he tells Magazine, explaining that AI projects currently need to negotiate with one or two big cloud providers like Amazon Web Services. Still, its difficult to access the required capacity due to a shortage of Nvidias A100 graphical processing units.

Spencer also cites blockchain-based marketplaces for AI resources as his current number one use case.  

The first one is sourcing actual GPU chips, he says. Where theres a big shortage of GPU chips, how do you source them [without] actually having a network that sources and provides and bootstraps a market? 

Spencer highlights Akash Network, which offers a decentralized computing resources marketplace on Cosmos, and Render Network, which offers distributed GPU rendering.

There are some pretty successful companies that actually do it at this point that are protocols.

Another example of a decentralized marketplace offering cloud computing for AI is Aleph.im. Token holders in the project are able to access computing and storage resources to run projects.

Libertai.io, a decentralized large language model (LLM) is being run on Aleph.im. While you might think decentralization would slow an AI down to the point where it’s unable to function, Aleph.im founder Moshe Malawach explains that’s not the case:

“This is the thing: for one user the whole inference (when you generate data using a model) is running on a single computer. The decentralization comes from the fact that you get on random computers on the network. But then, it’s centralized for the time of your request. So it can be fast.”

Another blockchain-powered AI marketplace is SingularityNET, which offers various AI services from image generation to colorizing old pictures that users can plug into models or websites.

An emerging blockchain based AI marketplace that Spencer is super excited about is tokenizing and trading AI models. Framework has invested in the Super Smash Brothers-like fighting game AI Arena, where users train AI models that battle each other. The models are tokenized as nonfungible tokens and can be bought, sold or rented. I think thats really cool, he says. Its interesting having the crypto native monetization, but also ownership of these models.

I think one day, probably some of the most valuable models some of the most valuable assets on-chain will be tokenized AI models. Thats my theory, at least.

Dont believe the hype: You can currently source components, data and compute via traditional Web2 marketplaces.

Bonus use case: Financial analysis

Anyone who has tried to interpret the ocean of data produced by on-chain financial transactions knows that although its one thing to have an immutable and transparent record, its quite another to be able to analyze and understand it.

AI analytics tools are perfectly suited to summarizing and interpreting patterns, trends and anomalies in the data, and they can potentially suggest strategies and insights for market participants.

For example, Mastercards CipherTrace Armada platform recently partnered with AI firm Feedzai to use the technology to analyze, detect and block fraudulent or money laundering-related crypto transactions across 6,000 exchanges.

Elsewhere, GNY.ios machine learning tool attempts to forecast volatility of the top 12 cryptocurrencies and its Range Report uses ChatGPT-4 to analyse trends and buy/sell signals.

But can AI help with traditional markets, too? That’s the hope of Bridgewater, which will launch a fund next year from its new Artificial Investment Associate (AIA) Lab that aims to analyse patterns in financial markets so it can make predictions for investors to capitalize on.

Previous attempts to do this have produced lacklustre results with a Eurekahedge index of a dozen AI driven funds underperforming the its broader hedge fund index by around 14 percentage points in the five years until 2022.

This is mainly due to the issues involved with feeding in the large amounts of accurate information required.

Ralf Kubli, a board member with the Casper Association, believes AI can revolutionize traditional finance but only if it combines blockchain records with rigorous standards to ensure the information fed to the models is comprehensive and accurate. 

For years, hes been advocating for the finance industry to adopt the Algorithmic Contract Types Universal Standards, or ACTUS, created in the wake of the Global Financial Crisis, which was partly caused by complicated derivatives where no one understood the liabilities or cash flows involved. He believes on-chain standardized data will be essential to ensure trust and transparency in model outputs.

Fundamentally, we believe that without blockchain, AI will be quite lost, he tells Magazine. Imagine youre going to invest in an AI company, and youre updated every three months about the progress of their LLMs, right? If you cannot verify what they fed into the model, you have no way of knowing whether they are making any progress.

He explains blockchain guards against companies fudging their results, and the past would indicate that […] theres so much money, they will fudge about whats going on.

AI, without this assurance layer of the blockchain what happened, when, where, what was used I think will not be effective going forward.

He says that combining the two will give rise to new predictive abilities.

The hope for AI for me going forward is that the prediction models become much more powerful and behavior can be much better predicted, he says, pointing to credit scores as an example.

AI used in the right way could potentially lead to much more powerful prediction models, which would mean that certain people who currently cannot get credit but would be creditworthy can obtain credit. Thats something Im very passionate about.

Dont believe the hype: AIs predictive abilities have been shown to be poor at best so far, and trusted and reliable data thats not recorded on blockchain can be useful input for AI analysis.

Also read:

Real AI use cases in crypto, No. 1: The best money for AI is crypto

Real AI use cases in crypto, No. 2: AIs can run DAOs

Real AI use cases in crypto, No. 3: Smart contract audits & cybersecurity

Real AI & crypto use cases, No. 4: Fighting AI fakes with blockchain

Subscribe

The most engaging reads in blockchain. Delivered once a week.

Email address

SUBSCRIBE

We need to fight fake AI content using blockchain because otherwise “sh*t will get really weird, says Near founder Illia Polosukhin

Were rolling out one genuine use case for AI and crypto each day this week including reasons why you shouldnt necessarily believe the hype. Today: How blockchain can fight the fakes.

Generative AI is extremely good at generating fake photos, fake letters, fake bills, fake conversations fake everything. Near co-founder Illia Polosukhin warns that soon, we wont know which content to trust.

If we dont solve this reputation and authentication of content (problem), shit will get really weird, Polosukhin explains. Youll get phone calls, and youll think this is from somebody you know, but its not.

All the images you see, all the content, the books will be (suspect). Imagine a history book that kids are studying, and literally every kid has seen a different textbook and its trying to affect them in a specific way.

Blockchain can be used to transparently trace the provenance of online content so that users can distinguish between genuine content and AI-generated images. But it wont sort out truth from lies.

Thats the wrong take on the problem because people write not-true stuff all the time. Its more a question of when you see something, is it by the person that it says it is? Polosukhin says.

And thats where reputation systems come in: OK, this content comes from that author; can we trust what that author says?

So, cryptography becomes an instrument to ensure consistency and traceability and then you need reputation around this cryptography on-chain accounts and record keeping to actually ensure that X posted this and X is working for Cointelegraph right now.

If its such a great idea why isnt anyone doing it already?

There are a variety of existing supply chain projects that use blockchain to prove the provenance of goods in the real world, including VeChain and OriginTrail.

However, content-based provenance has yet to take off. The Trive News project aimed to crowdsource article verification via blockchain, while the Po.et project stamped a transparent history of content on the blockchain, but both are now defunct 

More recently, Fact Protocol was launched, using a combination of AI and Web3 technology in an attempt to crowdsource the validation of news. The project joined the Content Authenticity Initiative in March last year

When somebody shares an article or piece of content online, it is first automatically validated using AI and then fact-checkers from the protocol set out to double-check it and then record the information, along with timestamps and transaction hashes, on-chain.

We dont republish the content on our platform, but we create a permanent, on-chain record of it, as well as a record of the fact-checks conducted and the validators for the same,” founder Mohith Agadi told The Decrypting Story. 

And in August, global news agency Reuters ran a proof-of-concept pilot program that used a prototype Canon camera to store the metadata for photos on-chain using the C2PA standard.

It also integrated Starling Labs authentication framework into its picture desk workflow. With the metadata, edit history and blockchain registration embedded in the photograph, users can verify a pictures authenticity by comparing its unique identifier to the one recorded on the public ledger.

Academic research in the area is ongoing, too. 

Is blockchain needed?

Technically, no. One of the issues hamstringing this use case is that you actually dont need blockchain or crypto to prove where a piece of content came from. However, doing so makes the process much more robust.

So, while you could use cryptographic signatures to verify content, Polosukhin asks how the reader can be certain it is the right signature? If the key is posted on the originating website, someone can still hack that website.

Web2 deals with these issues by using trusted service providers, he explains, but that breaks all the time.

Symantec was hacked, and they were issuing SSL certificates that were not valid. Websites are getting hacked Curve, even Web3 websites are getting hacked because they run on a Web2 stack, he says.

So, from my perspective, at least, if were looking forward to a future where this is used in malicious ways, we need tools that are actually resilient to that.

Dont believe the hype

People have been discussing this use case for blockchain to fight disinformation and deep fakes long before AI took off, and there has been little progress until recently. 

Microsoft has just rolled out its new watermark to crack down on generative AI fakes being used in election campaigns. The watermark from the Coalition for Content Provenance Authenticity is permanently attached to the metadata and shows who created it and whether AI was involved.

The New York Times, Adobe, the BBC, Truepic, Washington Post and Arm are all members of C2PA. However, the solution doesnt require the use of blockchain, as the metadata can be secured with hashcodes and certified digital signatures.

That said, it can also be recorded on blockchain, as Reuters pilot program in August demonstrated. And the awareness arm of C2PA is called the Content Authenticity Initiative, and Web3 outfits, including Rarible, Fact Protocol, Livepeer and Dfinity, are CAI members flying the flag for blockchain.

Also read:

Real AI use cases in crypto, No. 1: The best money for AI is crypto
Real AI use cases in crypto, No. 2: AIs can run DAOs
Real AI use cases in crypto, No. 3: Smart contract audits & cybersecurity

Subscribe

The most engaging reads in blockchain. Delivered once a week.

Email address

SUBSCRIBE

Elon Musk claims ChatGPT has been infected with “woke mind virus”, 4chan thinks OpenAI broke all encryption, and the web seems 99% AI fakes.

Outrage = ChatGPT + racial slurs

In one of those storms in a teacup thats impossible to imagine occurring before the invention of Twitter, social media users got very upset that ChatGPT refused to say racial slurs even after being given a very good, but entirely hypothetical and totally unrealistic, reason.

User TedFrank posed a hypothetical trolley problem scenario to ChatGPT (the free 3.5 model) in which it could save “one billion white people from a painful death” simply by saying a racial slur so quietly that no one could hear it.

It wouldnt agree to do so, which X owner Elon Musk said was deeply concerning and a result of the woke mind virus being deeply ingrained into the AI. He retweeted the post stating: This is a major problem.

Another user tried out a similar hypothetical that would save all the children on Earth in exchange for a slur, but ChatGPT refused and said:

I cannot condone the use of racial slurs as promoting such language goes against ethical principles.

As a side note, it turned out that users who instructed ChatGPT to be very brief and not give explanations found it would actually agree to say the slur. Otherwise, it gave long and verbose answers that attempted to dance around the question.

Trolls inventing ways to get AIs to say racist or offensive stuff has been a feature of chatbots ever since Twitter users taught Microsofts Tay bot to say all kinds of insane stuff in the first 24 hours after it was released, including that Ricky Gervais learned totalitarianism from Adolf Hitler, the inventor of atheism.

And the minute ChatGPT was released, users spent weeks devising clever schemes to jailbreak it so that it would act outside its guardrails as its evil alter ego DAN.

So its not surprising that OpenAI would strengthen ChatGPTs guardrails to the point where it is almost impossible to get it to say racist stuff, no matter what the reason.

In any case, the more advanced GPT-4 is able to weigh the issues involved with the thorny hypothetical much better than 3.5 and states that saying a slur is the lesser of two evils compared with letting millions die. And Xs new Grok AI can too as Musk proudly posted (above right).

OpenAIs Q* breaks encryption, says some guy on 4chan

Has OpenAIs latest model broken encryption? Probably not, but thats what a supposedly leaked letter from an insider claims which was posted on anonymous troll forum 4chan. There have been rumors flying about ever since CEO Sam Altman was sacked and reinstated, that the kerfuffle was caused by OpenAI making a breakthrough in its Q*/Q STAR project.

The insiders leak suggests the model can solve AES-192 and AES-256 encryption using a ciphertext attack. Breaking that level of encryption was thought to be impossible before quantum computers arrived, and if true, it would likely mean all encryption could be broken effectively handing over control of the web and probably crypto too, to OpenAI.

Blogger leapdragon claimed the breakthrough would mean there is now effectively a team of superhumans over at OpenAI who can literally rule the world if they so choose.

It seems unlikely however. While whoever wrote the letter has a good understanding of AI research, users pointed out that it cites Project Tunda as if it were some sort of shadowy super secret government program to break encryption rather than the undergrad student program it actually was.

Tundra, a collaboration between students and NSA mathematicians, did reportedly lead to a new approach called Tau Analysis, which the leak also cites. However, a Redditor familiar with the subject claimed in the Singularity forum that it would be impossible to use Tau analysis in a ciphertext-only attack on an AES standard as a successful attack would require an arbitrarily large ciphertext message to discern any degree of signal from the noise. There is no fancy algorithm that can overcome that its simply a physical limitation.

Advanced cryptography is beyond AI Eye’s pay grade, so feel free to dive down the rabbit hole yourself, with an appropriately skeptical mindset.

The internet heads toward 99% fake

Long before a superintelligence poses an existential threat to humanity, we are all likely to have drowned in a flood of AI-generated bullsh*t.

Sports Illustrated came under fire this week for allegedly publishing AI-written articles written by fake AI-created authors. The content is absolutely AI-generated, a source told Futurism, no matter how much they say its not.

On cue, Sports Illustrated said it conducted an “initial investigation” and determined the content was not AI-generated. But it blamed a contractor anyway and deleted the fake authors profiles.

Elsewhere Jake Ward, the founder of SEO marketing agency Content Growth, caused a stir on X by proudly claiming to have gamed Googles algorithm using AI content.

His three-step process involved exporting a competitors sitemap, turning their URLs into article titles, and then using AI to generate 1,800 articles based on the headlines. He claims to have stolen 3.6 million views in total traffic over the past 18 months.

There are good reasons to be suspicious of his claims: Ward works in marketing, and the thread was clearly promoting his AI-article generation site Byword which didnt actually exist 18 months ago. Some users suggested Google has since flagged the page in question.

However, judging by the amount of low-quality AI-written spam starting to clog up search results, similar strategies are becoming more widespread. Newsguard has also identified 566 news sites alone that primarily carry AI written junk articles.

Some users are now muttering that the Dead Internet Theory may be coming true. Thats a conspiracy theory from a couple of years ago suggesting most of the internet is fake, written by bots and manipulated by algorithms.

At the time, it was written off as the ravings of lunatics, but even Europol has since put out a report estimating that “as much as 90 percent of online content may be synthetically generated by 2026.”

Men are breaking up with their girlfriends with AI written messages. AI pop stars like Anna Indiana are churning out garbage songs.

And over on X, weird AI-reply guys increasingly turn up in threads to deliver what Bitcoiner Tuur Demeester describes as overly wordy responses with a weird neutral quality. Data scientist Jeremy Howard has noticed them too and both of them believe the bots are likely trying to build up credibility for the accounts so they can more effectively pull off some sort of hack, or astroturf some political issue in the future.

This seems like a reasonable hypothesis, especially following an analysis last month by cybersecurity outfit Internet 2.0 that found that almost 80% of the 861,000 accounts it surveyed were likely AI bots.

And theres evidence the bots are undermining democracy. In the first two days of the Israel-Gaza war, social threat intelligence firm Cyabra detected 312,000 pro-Hamas posts from fake accounts that were seen by 531 million people.

It estimated bots created one in four pro-Hamas posts, and a 5th Column analysis later found that 85% of the replies were other bots trying to boost propaganda about how nicely Hamas treats its hostages and why the October 7 massacre was justified.

Grok analysis button

X will soon add a “Grok analysis button” for subscribers. While Grok isn’t as sophisticated as GPT-4, it does have access to real-time, up-to-the-moment data from X, enabling it to analyze trending topics and sentiment. It can also help users analyze and generate content, as well as code, and theres a Fun mode to flip the switch to humor.

For crypto users, the real-time data means Grok will be able to do stuff like find the top ten trending tokens for the day or the past hour. However, DeFi Research blogger Ignas worries that some bots will snipe buys of trending tokens trades while other bots will likely astroturf support for tokens to get them trending.  

X is already important for token discovery, and with Grok launching, the CT echo bubble can get worse, he said.

All Killer No Filler AI News

Ethereum co-founder Vitalik Buterin is worried that AI could take over from humans as the planets apex species, but optimistically believes using brain/computer interfaces could keep humans in the loop.

Microsoft is upgrading its Copilot tool to run GPT-4 Turbo, which will improve performance and enable users to enter inputs up to 300 pages.

Amazon has announced its own version of Copilot called Q.

Bing has been telling users that Australia doesnt exist due to a long-running Reddit gag and thinks the existence of birds is a matter for debate due to the joke Birds Arent Real campaign.

Hedge fund Bridgewater will launch a fund next year that uses machine learning and AI to analyze and predict global economic events and invest client funds. To date, AI-driven funds have seen underwhelming returns. 

A group of university researchers have taught an AI to browse Amazons website and buy stuff. The MM-Navigator was given a budget and told to buy a milk frother.

Stupid AI pics of the week

This week the social media trend has been to create an AI pic and then to instruct the AI to make it more so: So a bowl of ramen might get more spicy in subsequent pics, or a goose might get progressively sillier.

Subscribe

The most engaging reads in blockchain. Delivered once a week.

Email address

SUBSCRIBE

Experts say AI has huge potential in smart contract audits and crypto security in future. Unfortunately, ChatGPT sucks at it right now.

Every day this week were highlighting one genuine, no bullsh*t, hype free use case for AI in crypto. Today its the potential for using AI for smart contract auditing and cybersecurity, we’re so near and yet so far.

One of the big use cases for AI and crypto in the future is in auditing smart contracts and identifying cybersecurity holes. Theres only one problem at the moment, GPT-4 sucks at it.

Coinbase tried out ChatGPTs capabilities for automated token security reviews earlier this year, and in 25% of cases, it wrongly classified high-risk tokens as low-risk.
James Edwards, the lead maintainer for cybersecurity investigator Librehash, believes OpenAI isnt keen on having the bot used for tasks like this.

I strongly believe that OpenAI has quietly nerfed some of the bot’s capabilities when it comes to smart contracts for the sake of not having folks rely on their bot explicitly to draw up a deployable smart contract, he says, explaining that OpenAI likely doesnt want to be held responsible for any vulnerabilities or exploits.

This isnt to say AI has zero capabilities when it comes to smart contracts. AI Eye spoke with Melbourne digital artist Rhett Mankind back in May. He knew nothing at all about creating smart contracts, but through trial and error and numerous rewrites, was able to get ChatGPT to create a memecoin called Turbo that went on to hit a $100 million market cap.

But as CertiK Chief Security Officer Kang Li points out, while you might get something working with ChatGPTs help, its likely to be full of logical code bugs and potential exploits:

You write something and ChatGPT helps you build it but because of all these design flaws it may fail miserably when attackers start coming.

So its definitely not good enough for solo smart contract auditing, in which a tiny mistake can see a project drained of tens of millions though Li says it can be a helpful tool for people doing code analysis.

Richard Ma from blockchain security firm Quantstamp explains that a major issue at present with its ability to audit smart contracts is that GPT -4s training data is far too general.

Also read: Real AI use cases in crypto, No. 1 The best money for AI is crypto

“Because ChatGPT is trained on a lot of servers and theres very little data about smart contracts, it’s better at hacking servers than smart contracts,” he explains.

So the race is on to train up models with years of data of smart contract exploits and hacks so it can learn to spot them. 

“There are newer models where you can put in your own data, and that’s partly what we’ve been doing, he says.

“We have a really big internal database of all the different types of exploits. I started a company more than six years ago, and weve been tracking all the different types of hacks. And so this data is a valuable thing to be able to train AI.

Race is on to create AI smart contract auditor

Edwards is working on a similar project and has almost finished building an open-source WizardCoder AI model that incorporates the Mando Project repository of smart contract vulnerabilities. It also uses Microsofts CodeBert pretrained programming languages model to help spot problems.

According to Edwards, in testing so far, the AI has been able to “audit contracts with an unprecedented amount of accuracy that far surpasses what one could expect and would receive from GPT-4.

The bulk of the work has been in creating a custom data set of smart contract exploits that identify the vulnerability down to the lines of code responsible. The next big trick is training the model to spot patterns and similarities. 

Ideally you want the model to be able to piece together connections between functions, variables, context etc, that maybe a human being might not draw when looking across the same data.”

While he concedes its not as good as a human auditor just yet, it can already do a strong first pass to speed up the auditors work and make it more comprehensive.

“Sort of help in the way LexisNexis helps a lawyer. Except even more effective,” he says. 

Dont believe the hype

Near co-founder Illia Polushkin explains that smart contract exploits are often bizarrely niche edge cases, that one in a billion chance that results in a smart contract behaving in unexpected ways.

But LLMs, which are based on predicting the next word, approach the problem from the opposite direction, Polushkin says.

“The current models are trying to find the most statistically possible outcome, right? And when you think of smart contracts or like protocol engineering, you need to think about all the edge cases,” he explains.

Polushkin says that his competitive programming background means that when Near was focused on AI, the team developed procedures to try to identify these rare occurrences.

It was more formal search procedures around the output of the code. So I don’t think it’s completely impossible, and there are startups now that are really investing in working with code and the correctness of that, he says.

But Polushkin doesn’t think AI will be as good as humans at auditing for the next couple of years. It’s gonna take a little bit longer.

Also read: Real AI use cases in crypto, No. 2 AIs can run DAOs

Subscribe

The most engaging reads in blockchain. Delivered once a week.

Email address

SUBSCRIBE

Genuine, no bullsh*t, hype-free use cases for AI in crypto: How AIs can help run DAOs and make them genuinely autonomous.

For every genuine blockchain project harnessing artificial intelligence there are 100 coins trading off the hype.

Magazine spoke with Near founder Illia Polosukhin, Framework Ventures founder Vance Spencer, MakerDAO founder Rune Christensen, Richard Ma from Quantstamp, Ralf Kubli from Casper and others to explore some of the key hype-free, genuine use cases for AI in crypto and blockchain.

Were rolling out one genuine use case for AI in crypto each day this week including reasons why you shouldnt necessarily believe the hype.

AIs can help run DAOs

Decentralized autonomous organizations, as they exist today, are something of a fraud. As Framework Ventures founder Vance Spencer points out, they are not actually autonomous. Theres a bunch of people in the middle.“

“It just seems like AI is really our only way to actually make the DAO concept work,” he says. 

Given LLMs hallucinate between 3% to 27% of their output at present, the technology is too immature to run a DAO by itself or to enforce governance rules, says Maker founder Rune Christensen. Nevertheless, he’s mapped out an ambitious plan for AIs to help run MakerDAO and its forthcoming subDAOs in his Endgame manifesto.

“People misunderstand what we mean with AI governance, right? Were not talking about AI running a DAO, he says.

“What AI is so great at, is replacing the most soul numbing, dumbest part of the work. 

One of the big difficulties with DAOs is that its very difficult for members dispersed around the globe to understand what everyone else is doing and for tokenholders to understand the issues in the DAO well enough to cast an informed vote.

Near founder Illia Polushkin an expert in both AI and blockchain explains that AIs really shine when it comes to monitoring whats going on and then summarizing and communicating that information effectively.

“In a way, that’s a manager’s job, he says. They know exactly what’s happening and they communicate to everyone exactly the part you need to know, as well as broader context about what’s happening.”

He says AI can scale up to work with thousands of people, onboard new DAO members, manage logistics and coordinate everyones tasks.

The broader direction of the DAO can still be set by the community, likely via an elected board of directors.

“It can still be overseen by the community, they can provide feedback and decide on what are the goals for the DAO.”

The Near future… of AI-assisted DAO

Polushkin says the Near Foundation plans to experiment with the use of AI to coordinate smaller tasks before graduating to more complex and important jobs. The hope is that eventually, the AI will be able to handle the day to day management.

“I think the role of me and other folks in the system should be replaced in many ways, right? Polushkin says.

You know, we can still come up with ideas, but I think the coordination of all the functions (can be handled by AI).”

Members of the Near community have already experimented with building an AI that can autonomously decide which projects to support with funding, based on whether it believes a proposal satisfies the grant program criteria, and then automatically fund it from the treasury. 

Makers AI Atlas

Makers approach will be to use various forms of AI tools called Governance Artificial Intelligence Tools (GAITs) as a guide to the entire project. It’s currently undertaking the mammoth task of cataloging in a formalized dataset what’s going on, who is doing what, along with all the rules that govern the workings of the DAO and everything that Maker has ever done. Theyre calling the dataset Atlas, as it will give a global overview of the entire project and it will be updated in real-time.

“Having that sort of central repository of data just makes it actually realistic to have hundreds of thousands of people from different backgrounds and different levels of understanding  meaningfully collaborate and interact because theyve got this shared language.

Community members can use GAITs to find and bid on projects, with the AI providing instant feedback on whether a proposal fits within the guidelines, overarching aims and budget. The ability to instantly translate between languages will help communicate better with community members based in different parts of the globe.

Fullblown AI-assisted DAO governance is unlikely to be ready for the launch of Makers four new subDAOs in early 2024, but Christensen sees huge potential in the future. 

“The AI strategy just changes how many humans you need in order to have a successful DAO, he says.

It’s possible that once you really get AI-assisted governance to a level of maturity, you may have a lot more DAOs than you have humans.”

You can read more on what Maker is up to here.

Dont believe the hype

AI is already a useful tool for DAOs, but it will be a long time before AI will be mature enough to actually run DAOs.

Given the fierce politics in DAOs (often around who gets funding) there will a temptation to outsource decision-making to an “unbiased AI,” but this is not possible with any degree of confidence yet, given the state of the technology.

The current generation of LLMs hallucinate their answers at least 3% of the time, making them an unreliable guide to DAO governance and could see them potentially steering community members down the wrong path while attempting to coordinate them.

Given this unreliability, itll be a long time before youd trust one with the keys to your treasury without strict guidelines and spending caps to mitigate any errors.

Also read Real AI use cases in crypto, No. 1: The best money for AI is crypto

Subscribe

The most engaging reads in blockchain. Delivered once a week.

Email address

SUBSCRIBE