Is every organisation ready for implementing and really benefitting from AI? Are all industries equally poised to succeed? What are the core use cases? Such considerations are common even among those who have already started their journey towards AI-enabled enterprise. In this December blog entry, we will share our views and experiences - maybe they could serve as an inspiration for some New Year’s resolutions ;)
According to McKinsey, the top 5 industries most impacted by AI value creation (read: potentially reaping most benefits from AI implementations) are:
- transport & logistics
- automotive & assembly
- high tech
A revered venture capital firm - MMC Ventures - lists 31 core AI use cases in eight sectors:
- asset management
- law & compliance
The lists don’t really overlap, do they? This alone shows that there is no common consent as to who is going to benefit most from the implementation of AI - maybe because there is no sufficient research data and the topic so dynamic.
The very same McKinsey claims that in 60% of occupations at least 30% of constituent activities are technically automatable by adapting proven AI technologies. So, while there are some industries which could probably benefit more than others, the AI is basically needed everywhere. Brainhint customers so far recruit at least from the following industries (the list expands every day): apparel/retail, automotive, construction materials, electronics, finance, food packaging, industrial machinery, media, pharma, retail. And all of them enjoyed tangible benefits from implementations – more on that later.
Decisively, China is well ahead of the game in AI adoption, which to some may seem surprising, as it is tempting to think that Silicon Valley and in general United States have dominated the field. Well, in development of core technology they might lead the pack, but as far as adoption rate by general economy is concerned, North America, not mentioning Europe, is lagging behind. Reasons for that might be that the organisations in China:
- enjoy strong support from government, including less-stringent data protection policies (especially compared to Europe),
- exhibit advanced approach to corporate data (MIT Sloan claims 78% of leading Chinese companies maintain their data in centralised data lake, compared with 43% in the US and 37% in Europe),
- have fewer legacy systems.
The European AI ecosystem needs to catch-up, with the policymakers not really making it easier (on the other hand, as citizens we should probably be happy about the privacy level we can enjoy compared with our Chinese counterparts). But does it mean we have lost the game already? We don’t think so.
So then - is it industry, location on the globe or rather leadership maturity and mindset?
Thanks to huge developments in technology, typically human tasks like understanding, reasoning and planning can nowadays be performed by AI quite efficiently and at reasonable expense which makes sense among most industries all-over the world. We believe at Brainhint that it is about leaders’ attitude whether their organisations will be among frontrunners or laggards in the AI implementation race. Organisations considering “if” are quickly outpaced by those contriving details “how” and the gap between the latter and the first then expands rapidly. Once the organisations start investing in AI, they quickly increase budgets, as they see improves in productivity, revenues and earnings.
AI Maturity Curve
A simple AI Maturity Curve, proposed this year by EY (commissioned by Microsoft), includes 5 levels:
- None – not yet thinking about AI
- Planned – AI is being planned but not yet put to active use, not even in early stage
- Piloting – AI is put to active use, but still only in early stage pilots
- Released – AI is put to active use in one or a few processes in the company but still quite selectively, and/or not enabling very advanced tasks
- Advanced – AI is actively contributing to many processes in the company and is enabling quite advanced tasks
with majority of companies across 15 European markets in Piloting (41%) or Released (27%) phase. Wherever your organisation might be on this curve (hopefully at least in the middle), it is crucial not only to push forward but to work hard on the clarity of each business case - implementation of AI for the sake of AI is a straight path to failure or at least discouragement.
Measures of success
At Brainhint, we focus with all our customers on exact definition of business case each time we talk about potential application of AI in their organisations. And that is why we believe there is no single right methodology of measuring success other than time saved or money earned. Be it cost per click decrease (superauto24.pl, minus 7%), operational effectiveness of debt portfolio (Statima, +36%), automation of a process (Siemens, 100% automation of RFQ verification) and more.
Analysis, goal, measures and feedback – the shell where AI feels best – are a must. This is why we put so much pressure on recognizing what lies beneath the needs of our customers, assigning them dedicated enterprise advisor instead of just a development team. The perfect AI needs a smart human to get the job done.
you may also like