Why rushing into AI is not the answer to your business problems

You could say 2022 was the year of Artificial Intelligence.?

OpenAI launched ChatGPT3 in November last year, and suddenly the media sat up.?

The attention shocked even the developers. Though the largest language model (LLM) ever created, it launched with little fanfare. Development had been going on for years and to a certain extent, the wider public wasn’t interested. Then everyone took notice.?

Generative AI burst onto the scene in a big way. People shared viral screenshots of strangely convincing AI-generated text and their increasingly creative prompts. Fantastical AI selfies littered Instagram feeds and headlines filled with concerns from business leaders, educators and artists about what it all meant.

Cut to large businesses, and AI had already been a hot topic for years. AI was already a huge talking point in enterprise tech. With its promises of unparalleled efficiency, improved decision-making, and streamlined operations, AI had captured the imagination of executives seeking ways to drive growth and stay ahead of the competition.?

But after generative AI burst onto the scene, suddenly individuals and smaller businesses were envisaging their own everyday applications of AI. A note of panic entered business communities about what or whose roles AI would make obsolete, and businesses of all sizes were realising that they didn’t want to be among the stragglers left in the dust.

Today AI hype is at new heights, and many fear that they will be left behind. At the same time, the financial pressures facing businesses mean business leaders are being pressed to deliver on growth. In this environment, executives might find themselves tempted to rush into AI investments without considering the underlying complexities and implications.

It’s clear to me as a Knowledge Management and tech expert that there is no reason for AI ‘fear of missing out’. Above all, businesses have to avoid rushing into investments in their eagerness to embrace AI without proper ROI research. Instead, they should look at the lessons early adopters of AI enterprise tech have already learned: that whether you get a return on your investment in an AI implementation depends on how well you prepare, how good your data is, and appropriate use cases. Reports have always shown that ROI from AI investment varies wildly. Like any disruptive technology, it takes a degree of maturity before it pays for itself.?

AI is a tool, not a magic wand. Getting those returns depends on how well you do your organisational transformation before you bring in the AI, and then how well you scale it.?

The struggle to scale

Although the power of emerging AI technologies are undeniable, the challenge of validating the business use cases, implementing them, and most importantly scaling up to get the value of investments back in good time is still proving to be too much of a challenge for most enterprises.

Despite eye-watering investment from big business such as Google with their DuetAI, Microsoft with their Co-pilots and now Apple with AppleGPT on its way, AI is still an emerging technology, or rather technologies. Not all necessarily at a stage of being able to serve real business needs in a reliable business model.?

The number of AI startups solving all possible enterprise use cases grows on a daily basis. The most frequent one is the AI magic wand called “the ultimate search capability which will retrieve all knowledge one needs from multiple data sources”. The promise of the top smart knowledge retrieval layer is there, however enterprises need to address their real core problem first - to structure their in-house data and implement appropriate content and knowledge management strategies.?

A report published this year by 埃森哲 found that nearly three-fourths (73%) of companies are prioritising AI over all other digital investments. The same research found that only 12% of the companies experimenting with enterprise AI are using it at a maturity level that achieves a strong competitive advantage.

Gartner , Inc.’s annual global survey of CIOs and technology executives emphasised the increasing urgency for CIOs and IT leaders to accelerate time to value from their AI investments. Citing inflation, talent scarcity and supply challenges, the survey highlighted the importance of driving bottom-line growth, and importantly, getting there faster.

Business leaders hoping to speed up time to value will want to keep in mind the above mentioned research from Accenture, which showed an eye-watering ROI gap between businesses depending on the maturity of their AI implementation. Companies that were “strategically scaling AI report nearly 3X the return from AI investments compared to companies pursuing siloed proof of concepts”.

The numbers speak for themselves. If you’re not on board with the organisational transformation, or a full scale implementation then save your money. Dabbling in enterprise AI won’t pay for itself. At least not yet.

The role of Knowledge Management in AI success

As a Knowledge Management and tech expert, I have witnessed first hand the transformative power of AI when complemented by a comprehensive knowledge management and data strategy.?

There’s no doubt AI is revolutionary, and offers forward-thinking businesses enormous potential value. But without a carefully crafted strategy, a detailed evaluation of existing enterprise architecture, data and knowledge flows, and information security, businesses risk falling into common pitfalls, such as irrelevant implementations, poor adoption, siloed efforts and data concerns. Or in most cases, the wrong output and lowered expectations of AI value, (AKA disappointment).?

Knowledge Management serves as the backbone of successful AI implementation by addressing critical factors, such as:

• Identifying relevant use-cases

Knowledge management allows businesses to thoroughly assess their operations and identify areas where AI can bring meaningful value. This ensures that AI investments align with the organisation's objectives and contribute to its growth, rather than exist in silos with difficulties to measure.

• Preparing data and ensuring quality

Effective knowledge management includes robust data governance practices, ensuring that data is standardised, cleaned, and ready for AI applications. This enhances the accuracy and reliability of AI-generated insights.

• Building human-AI collaboration

Knowledge management facilitates the integration of AI into existing workflows, ensuring seamless collaboration between employees and AI systems. Proper training and upskilling of the workforce foster a culture of acceptance and utilisation of AI tools. Making human-AI collaboration a habit and a norm is a must.

• Scalable transformation

An effective knowledge management strategy lays the groundwork for a scalable AI transformation. It allows businesses to adopt AI in a phased manner, allowing for iterative improvements, ensuring AI agility, and avoiding costly disruptions. With access to enterprise content and data, knowledge management can help to identify additional applications of AI solutions and drive cross-departmental adoption.?

The take home message?

Enterprises can future-proof themselves with large scale purpose built solutions, and now smaller and medium sized businesses can benefit enormously from free and subscription-based AI tools for enhanced efficiency too. But, in the pursuit of growth and navigating financial pressures, businesses must resist hasty AI investments and instead focus on their own well-planned strategy that serves their organisation. AI is complementary, not substitutional.?

Other businesses may have already invested in AI long ago but it doesn’t mean they are yet seeing ROI. So don’t fall into FOMO! Successful AI adoption requires a thoughtful and scalable transformation approach, coupled with robust knowledge management practices and data management strategy. By identifying relevant use-cases, preparing data, fostering human-AI collaboration, and ensuring scalability, organisations can unleash AI's true potential and achieve desired ROI.?

The truth is, getting roles, responsibilities, change management and culture right is just as crucial as the process and tech. Ultimately, business value success lies not in the most advanced technology but in how well it's evaluated, implemented, adopted, and scaled. AI tools can only reach their full potential when guided by human intelligence, not the other way around.