Help Me Democratize AI!

Artificial Intelligence is a major technology catalyzing innovation across virtually all industries, but it's not being done fairly.

Most of current AI advancements fall in the Machine Learning bucket, and more specifically, "Deep Learning," a form of deep neural networks that uses a lot of data. While AI is largely a hype technology, and can be described as "dynamically optimized statistical techniques," rebranded, it is a massive industry in itself.

It's also extremely difficult to implement in the real-world, which is why 85% of big data projects fail (according to Gartner), even though a Deloitte survey of “aggressive adopters” of cognitive technologies found that 76% believe that they will “substantially transform” their companies within the next three years.

Most companies never move beyond the proof-of-concept phase in AI, if they're lucky enough to get there in the first place. And yet, AI giants are taking over entire industries that fail to stay ahead of the curve.

It's due to a few reasons:
• Difficulty finding AI talent
• Budget constraints
• Data issues and disparate data sources
• Lack of understanding of AI

Huge companies hoard the data, talent, and models needed to fuel AI, so the value isn't equitably distributed. Platforms like Steemit aim to solve this problem for specific use-cases (e.g. social media).

My company, bitgrit, looks at solving those problems by merging artificial intelligence with decentralization technologies, namely blockchain.

Now, you might be getting skeptical when I namedrop both "AI" and "blockchain" as part of our solution mechanism, but let me explain.

Being able to trust AI requires transparency and auditability in the sources of data that are used to train it, requiring open cooperation between the providers of training data, data scientists, and end users. These needs can be met by the immutable frameworks for trust provided by distributed ledger technology.

Besides trust in AI systems, we want to make it easier for data scientists to work on real-world problems.

To that end, we've designed two main services, a Data Science Network (think Tinder for data scientists), and an AI Marketplace (Uber-esque matchmaking of data talent and solutions). The data science network is built around a growing community - our most active is on http://t.me/bitgrit, with around 850 data scientists, and our forum and job board are live at forum.bitgrit.net. The marketplace is under development, and the core of our vision to ensure that the value from AI is transmitted transparently and equitably to the people responsible for creating it.

If this sounds like your cup of tea, spread the word among data scientists, AI geeks, and really anyone interested in a fairer way of doing things. The easiest is to share our main community handle - http://t.me/bitgrit.