As one of the latest trends in the tech world, artificial intelligence (AI) is expanding into a variety of industries in new ways all the time. Companies are rushing to implement AI for many uses including high-frequency trading, monitoring social media presence, monitoring security, and autonomous vehicles.
Private companies aren’t alone in their pursuit of AI either; in recent news, the U.S. Pentagon announced its intentions to invest $2 billion into AI. But with all this interest in the emerging world of AI, one of the bottlenecks holding innovation in the industry back is the lack of affordable access to computing power.
AI’s Need for Computing Power
Computing power plays a major role for AI and is crucial to the continued development in the industry. In May of 2018, OpenAI published an analysis on AI and compute power and named 3 factors as main drivers in the advancing of AI: “algorithmic innovation, data (which can be either supervised data or interactive environments), and the amount of compute available for training.”
The amount of compute power utilized in AI development has grown exponentially in recent years and OpenAI’s analysis shows that since 2012:
“The amount of compute used in the largest AI training runs has been increasing exponentially with a 3.5 month-doubling time (by comparison, Moore’s Law had an 18-month doubling period). Since 2012, this metric has grown by more than 300,000x (and 18-month doubling period would yield only a 12x increase). Improvements in compute have been a key component of AI progress, so as long as this trend continues, it’s worth preparing for the implications of systems far outside today’s capabilities.”
With the large compute appetite in the AI industry, advancements are likely only able to continue with access to enough power for those processing AI. However, accessing hardware for handling AI development is not a cheap endeavor.
Right now, both new startups and household names in the tech industry are working to create AI-specific chips, but testing, manufacturing, and bringing these chips to market takes time and purchasing new hardware is expensive, especially when constructing the large infrastructure required for AI.
Fortunately, another major trend in the tech world is looking to bridge the gap between AI development and access to compute power: blockchain.
A Blockchain-Based Solution
To facilitate the growth of emerging AI technology, some are looking to another recent disruptive movement, blockchain technology, for solutions. Besides its implementations in the fintech world and storing data on an immutable ledger, the basis for blockchain technology (and the cryptocurrencies it sparked) is on decentralized networks.
Now, some are looking at these decentralized networks to offer the AI space access to more compute power in a more efficient manner than traditional centralized systems. Rather than forcing companies to invest in expensive hardware to develop AI on their own, blockchain-based approaches are looking to leverage a decentralized network of compute power that can be accessed by anyone.
One example of this is the blockchain-based startup Tatau which is creating the infrastructure for businesses and developers to connect with parties who have spare compute power. Tatau is building a platform to connect the two parties together (users and providers) to increase the efficiency of AI development.
Unlike traditional centralized approaches to cloud computing, a decentralized network inherently doesn’t have the same pricing structure. There’s no initial investment made purchasing new hardware that then requires a specific return on investment (ROI) with slim margins. Instead, providers sign up to utilize hardware they already own with spare compute power and capitalize on it when it’s not already in use.
With a decentralized model, users gain access to a more streamlined approach without the expensive overhead and keep prices at an accessible level for developers. This same decentralized approach has already disrupted other industries too.
While companies like Uber and Lyft are still centralized in ownership, the service providers are decentralized and, because of that, the price of an average Uber or Lyft ends up more accessible for consumers. When the service connecting two parties is not simultaneously the provider, we’ve seen a decrease in price for the consumer time and time again.
Similarly, look to a content provider like Alphabet’s YouTube. Though the platform is now introducing company-produced content, the basis of the platform has been on user-created content that didn’t require large investments from the platform to create. Instead, YouTube has been able to focus on connecting the two parties (creators and viewers) together and, as a result, has been able to do so without requiring the same payment structure as traditional television networks.
Tatau isn’t the only one in the compute space emulating this model either. Other blockchain-based startups are creating similar networks to tackle various tasks that require expensive hardware. New startups like Render Token and Leonardo Render are creating similar decentralized ecosystems for the creative industry to gain access to more rendering power.
To AI and Beyond
Many of the advancements in the tech world are coming from a convergence of multiple innovators and disruptors to make the barriers to advancement more accessible.
Using blockchain technology as a means for enabling the sharing economy is opening new doors to innovation and advancement that were previously too heavy for small businesses and entrepreneurs to open. While AI is still mainly at its early stages, it’s likely that blockchain could prove pivotal to its expansion and further development.
Disclaimer: This is a contributed article and should not be taken as investment advice