With AI, Forget The Killbots, and Follow The Money
Right now, just over half of enterprises globally say they are incorporating AI applications into their operations. According to McKinsey & Co, 70 per cent of companies will be using some form of AI in their business by 2030.
All those tasty price signals have attracted the attention of entrepreneurs, and coming in behind them, investors.
According to Venturebeat, AI investment reached $US77.5bn last year, almost double the 2020 levels. Like other parts of the tech sector AI and Machine Learning (collectively AI/ML) businesses are riding a wave of vast and accelerating digitalisation.
Artificial Intelligence has been a subject of fascination and close study, by both scientists and science-fiction writers, for decades. As long as there have been computers, we’ve wondered how close they could ever come to emulating human thought.
It doesn’t usually end well, at least not in the movies.
The reality of AI toasters wanting to murder their human masters is (at the current stage of technological development) not viable. It’s also a pretty poor business model for entrepreneurs looking to attract investors – although Elon Musk is apparently investing millions on developers to help him break out of the matrix.
Here in the world of real things, though, AI is already massive and more ubiquitous than many people realise and is being directed to both old and new world business problems. We explored further in NRP Report looking at four regional companies that are solving problems from honey bees to heavenly bodies and account receivables.
The one thing computers really can do better than humans is retain and process enormous amounts of data very, very quickly. To a computer, processing five data points is little different to processing five billion, although the more data you have, the more likely you can tease out patterns from it.
The more those patterns can enable you to make predictions –and the more often those predictions are right – the more that processing looks like intelligence. That’s the ugly secret of AI that the headlines ignore; it is less Terminator 2 and much more huge amounts of nested If/Then statements powered by massive and ever more cost-effective computing cycles – thank you Moore’s Law.
Think about weather prediction. The more we know about weather patterns, what’s happened in the past and what’s happening now, the more likely we are to guess right about what will happen tomorrow. The same is true for consumers’ shopping habits, supply chain management, or even certain medical applications. The more data you have, the more intelligent you seem.
Recent advances in processing power – plus growth in massive data centres, cloud-based services, and a proliferation of data collected from individuals and enterprises – have led to a boom in applications analysing and making sense of all that data. Businesses are investing in everything from natural language processing for customer service chatbots to self-driving cars to make use of such systems.
Indeed, many see AI as not only an enabler of business advantage, but a necessity for business survival. An Accenture survey found that 84 per cent of executives felt that AI was necessary for business advantage, and a whopping 75 per cent were worried that if they didn’t leverage AI within the next five years they would risk losing out to competitors.
The advantages of AI are of course compounding. Each insight generated by analytics becomes itself a data point to be added to the pool, to feed further insights.
Little wonder then that CB Insights cited a 108 per cent jump in funding for AI projects between 2020 and 2021, up to $66.8bn. GlobalData announced recently that February 2022 alone saw another $1.5bn worth of deals. Just as the pandemic has been great for cloud service providers, it’s been a boon for AI.
Intriguingly, CB Insights claims that 18 per cent of that funding – just over $12bn – was directed toward the healthcare industry. Retail, where everyone needs to know where the customer is going next, also saw huge growth in AI investment last year.
There were, according to CB Insights, 65 startup companies that reached a billion-dollar valuation in AI-related businesses through 2021. That’s a record, by a very long way. Indeed, it makes one wonder whether the term “unicorn” – used to describe a billion-dollar startup because they’re so rare – is even the right word to use when it comes to AI these days.
Expect further growth as the concept of the “metaverse” – an AI and VR-driven cyberscape on which Facebook founder Mark Zuckerberg, among others, has bet the future – starts to become reality. (Or, in our view, several different realities…)
Of course it’s not all unicorns and rainbows – there are also risks involved.
The biggest of these is trust. AI relies on massive amounts of data being available in order to work, and that data relies on people and organisations being willing to hand it over. Scandals such as Cambridge Analytica and other instances of misuse of data have led to a loss of confidence in the companies holding onto information about people.
This means that any company embarking on an AI project needs to be mindful – and communicative – about the way it collects data, the way it stores data (security and privacy have become hot-button topics) and the way it uses data.
Governments around the world have been slow to respond to such challenges with legislation, but they have responded. Numerous frameworks are in place, and adapting to a rapidly evolving technological environment. Enterprises need to design compliance with legal frameworks into their AI projects at the outset, including robust security to protect user data.
Smart computers make mistakes, mostly as a function of human frailty, and bias. AI projects (especially critical ones such as in healthcare and autonomous vehicles) must carefully navigate the practices and protocols of ethics in AI.
Indeed there’s probably a whole new investment opportunity in legal firms that specialise in pursuing the damages caused by algorithmic bias!
—
Footnote: North Ridge Partners and the firm’s founder Roger Sharp are proud to be working pro bono to help establish a more robust tech industry in Roger’s hometown of Queenstown, New Zealand. This is part of an effort to diversify the local economy away from its over reliance on the tourism sector. The relevance to this article?
In tech, everything starts with talent. So the first initiative conceived and launched in Queenstown is the Machine Learning Institute, a trade-level course that teaches school leavers and people who want to retrain, in basic AI/ML skills.
We will report more comprehensively on this during the year…