AI transforming retail
AI (Artificial Intelligence) is fundamentally technology enabling computer algorithms to mimic the human brain’s capacity to think, with the ultimate intention being to develop processing capabilities by far exceeding those of any human. The idea is that rather than having to be fed knowledge, AI-assisted algorithms are designed to grow smarter through observation and analysis, and to exponentially improve in effectiveness and efficiency by learning from mistakes and experience.
Online retail is one of the most exciting frontiers set to benefit from AI. Offering what is often no more than a searchable database, online retail has been known to deliver “hit or miss” product listings, leaving it up to shoppers to do all the refining. Artificial intelligence is bound to change all that, with AI set to power increasingly intuitive retail sites that make it easier and faster for shoppers to get to the products they desire, and generate increased conversion rates.
The technology will ultimately enable online stores to engage consumers much like talented sales representatives, who know everything there is to know about the products on offer, yet also develop and continuously enhance a keen understanding of individual shoppers’ personal preferences, style and sensibilities.
Employment landscape consequences
The impending value and potential of artificial intelligence is resulting in heightened demand for AI-specific talent in a range of industries, and retail is no different. AI is creating new jobs, with two out of every three them being senior level.
The biggest tech companies are ready to make significant investments in this field, whether as the means to reduce time spent on repetitive tasks, make complex jobs easier or increase sales. They’re consequently only too keen to throw big bucks in salaries and company shares on Ph.Ds. having just completed their studies or even fresh talent with only a limited track record, so long as they specialize in artificial intelligence research or development. Consider, for example, British AI lab DeepMind, which was acquired along with its 50 employees by Google for a reported $650 million in 2014. As per the company’s publicly released annual financial accounts in Britain, having grown to 400 employees by 2016, the lab’s staff costs amounted to $138 million, which translates to $345,000 per employee.
At the very top of the totem pole sit executives with AI project management experience, who’re paid as much as millions in compensation and other incentives to lead artificial intelligence development teams.
Is investing big bucks in in-house AI a recipe for success?
Well, not necessarily. With all the eagerness to invest in AI, it’s sometimes easy to forget just how hard, expensive and risky it can be to build AI systems that actually work. Even when backed by AI frameworks, artificial intelligence requires a level of knowledge and expertise by far exceeding that of mainstream Web developers, for example, and the number of people capable of building working AI prototypes is extremely limited.
Even the extreme scope of AI investments made tech giants, such as Google, can serve as a warning beacon of sorts. It should make it clear that this is a challenging and sometimes risky endeavor that’s not to be taken lightly or undertaken by those with limited financial or technological resources. Given the difficulty and rarity of achieving a leadership position in AI, the majority of companies (retailers included) is likely to end up in the AI passenger seat, with proprietary approaches made obsolete, and proven core AI technology ultimately provided by major industry players as part of open-source frameworks.
Retailers, for whom technology is auxiliary to their main line of business as it is, are particularly likely to benefit more from outsourcing AI. This will help them avoid throwing tons of money and resources on homegrown efforts, and maintain strong focus on their own core competencies.
And that’s a pretty surefire way of staying ahead of the technological and competitive curve, without overextending oneself.