Trends | 04-Jul-2017

AI-first to AI-standard

We are fast approaching the day when AI powers all the software we write. The idea of AI as separate to a technology solution will become inconceivable. It is therefore remarkable the number of companies, new and old, forming business plans and marketing campaigns around their AI.

There is a surfeit of companies coming to market with claims of AI emboldened solutions. For the most part, these companies are either capitalising on the current corporate landscape, which is subjugated by confounded executives or they have pioneered new AI techniques and packaged them for prospective application.

It is not the meaningful application of AI that is under scrutiny, there is little question of the value AI offers. The concern is on the basis of companies using AI as the differentiator and not the enablement mechanism.

Amazon recommends products based on machine learning algorithms, Netflix does the same with video entertainment and Spotify curates personalised playlists that are, in my own subjective experience, meticulous.

Could you imagine the world if Netflix had started as an “AI company" on the premise of developing recommendations engines (a world devoid of House of Cards, instead filled with productive Sundays) for movies and series based on your Blockbuster account order history? If Spotify had done similar with iTunes downloads, or CD purchases, for recommending new artists? The focus should be on solving the problem and then embracing AI to optimise, build and execute on that.

Retail companies such as Stitch Fix, Asos and Thread use similar AI techniques in product development, recommendations, inventory management and more. By using AI in meaningful ways they are continually improving aspects of the user experience, reducing wastage, aiding creative departments and adding value to their bottom line.

The development of state of the art AI techniques, once a domain reserved for a select few individuals chiefly in the academic field, are becoming commonplace in both consumer and enterprise technology. Software such as Google’s TensorFlow, Apache’s Hadoop and other open source libraries are democratising AI. These solutions drive the communities on which they thrive as the technology becomes more accessible and in the process propels innovation, enabling better solutions to be built within new and existing industries. 

Once mainstream adoption occurs AI will cease to be used as a marketing strategy or a genuine business plan. As ubiquity transpires across industries the idea of being an AI-first or AI-led company will gradually fade and AI will become a standard expectation, just as we have seen with relational databases, web-based applications and cloud computing. There will always be something shiny in technology, and while AI has the potential to be the greatest paradigm shifting advancement of this generation, it will likely give rise to something else.

The real value stems from the ability to identify areas of application, define the strategy and then in executing. The smart money is not on the next best dressed AI startup. The smart money is on utilising the right technologies to create world class products - to do so requires investment in the best people and a technology partner that works collaboratively to understand your market, identify the best solutions and then drives those to delivery.