The last six months have been a breakthrough for artificial intelligence (AI) development. Image processors like Mid journey and text generators like ChatGPT have sparked great public fascination and discussion. Contrary to the promise of driverless cars, these apps seem ready for real commercial use. That could pose a problem for technical innovators focused on public blockchains and distributed computing. At the very least, artificial intelligence will be a strong contender for investment funding for the foreseeable future, especially given the damage done to reputations by various scammers. And scams inflicted on cryptocurrencies in the last year.
There are also bigger differences between the underlying structure and ethos of blockchain and artificial intelligence. Previously, AI required large, centralized data caches for training, which could help create the kind of privacy and security risks that cryptocurrencies face. And the blockchain communities are deeply opposed. Additionally, this reliance on data caches renders AI ripe for capture by large, centralized enterprises. We’re seeing this right now: Microsoft’s $10 billion mammoth bet on Open AI and ChatGPT is clearly based on the restrictive use of the technology in enterprise products.
But there is a very different approach to AI that could help address this inherent anti-democratic bias. You see, apart from digital money, one of the clearest viable uses of blockchain is in the management of distributed computing resources. Blockchains can be used to coordinate and verify services on a network, with tokens providing incentives for taxpayers. So far, blockchains have been set up to power and govern things like distributed cloud storage (Filecoin) and high-end distributed graphics computing (Render Network). Another interesting example is BitTorrent, which started out as a non-tokenized peer-to-peer file-sharing network.
But researchers began to theories around 2017 that tokenizing the network could improve the network for all users by rewarding users with the fastest connections (see below for the technical details of the BTT implementation). Several blockchain projects have extended the logic of tokenized distributed computing to artificial intelligence. The most significant is probably SingularityNET, launched in 2017 by veteran AI researcher Ben Goertzel. Goertzel has been working on AI since the late 1980s, is the author of more than a dozen scholarly books on the subject, and is credited with popularising the term “artificial general intelligence.”
SingularityNET is one of the relatively few serious projects to raise funds during the ICO (initial coin offering) craze of 2017. Its token, AGIX, saw a huge rally following the deal between Microsoft and Open AI. Goertzel sums up SingularityNET’s often-together approach to artificial intelligence in the words of pioneering computer science researcher Marvin Minsky, who envisioned the development of AI as “a society of minds,” image generators that are good at certain things.
A SingularityNET user can request a specific combination of services on the network, which can be created and hosted anywhere in the world and freely associated with the network. This is basically similar to the way financial services can be bundled into larger packages called “compoundability” on a smart contract platform like Ethereum. However, Goertzel goes further and argues that a combinatorial environment for AI development, in which each module is developed independently, could be a better route to artificial general intelligence, i.e., human-like digital heads. A technical detail worth emphasising here is that artificial intelligence cannot run “on” a blockchain.
At least given today’s technology, this would be impossibly slow and expensive. SingularityNET and similar projects are dedicated to using blockchain to manage off-chain resources, which involves some technical risk as it requires layers of verification to ensure good inputs and outputs to and from the network.
Crypto philosophy vs. AI
Goertzel’s approach to AI is a useful contrast to current mainstream AI figurehead Sam Altman’s implicit beliefs. Altman is the CEO and co-founder of OpenAI, but he is also behind Worldcoin, a project that aims to create digital identities for world citizens by scanning their irises. On the face of it, Altman’s Worldcoin project is despicable for its willingness to collect incredibly sensitive biometrics from vulnerable populations. But its implied relationship with the AI could be even darker. Many social and economic theorists predict that AI will create a world of extreme inequality.
It is assumed that this will make workers such as truckers and cashiers obsolete. And because AI is more centrally controlled, those salaries become profits for big companies. Sam Altman’s Worldcoin is specifically designed to implement the main solution Silicon Valley has offered to this conundrum: universal basic income (UBI). The idea is that once the AI eliminates everyone’s jobs, some government or similar entity must tax the few centralized administrators of the AI with a tax high enough to redistribute the wealth to the common masses. Worldcoin’s eyeball-scanningsphere is primarily intended to ensure that nobody can fool this future system by collecting multiple payments.
I’m leftist and believe that some degree of redistribution of wealth makes society more stable and productive. But even to me, this UBI vision of Silicon Valley is an authoritarian nightmare that combines the worst attributes of bloated politburo communism and greedy monopoly capitalism. It would be a future in which many depend entirely on the generosity of a few. A techno-elite would run some sort of AI-powered shop of the exploitative kind that wails in traditional coalman’s ballads.
This is the future that the likes of Sam Altman and Peter Thiel are pointing out to us with obvious enthusiasm. While not yet a comprehensive alternative vision, distributed AI projects like SingluarityNET at least offer the hint of a more democratic technological future.