Atomic 212º recently held an event in partnership with LinkedIn and Microsoft to explore how AI is currently being applied in the B2B landscape. Atomic 212º CEO, Claire Fenner, breaks down the key findings and challenges discussed on the day.

AI is, almost painfully, the acronym on everyone’s lips right now. But while everyone is aware of it – and almost everyone is keen to reap some sort of benefit from it – when it comes to actually implementing AI in an organisation, a seemingly disproportionate amount of the responsibility is falling squarely on marketers.

Experimenting with AI applications to determine how companies can best work to drive more effective and efficient outcomes for unique businesses, customers and goals is a Herculean task.

Regardless of whether CMOs see this as an undue burden or opportunity to step up and own a whole new aspect of business operations, the true challenge that marketers now face is simply making time to actually deal with the issue head on. The hardest part is knowing where to start.

When AI is done right, the results can be game changing. Some interesting applications I have seen include the food giant Mars leveraging behavioural-based advertising research data, including eye tracking, EEG, facial reactions and more, to link specific behavioural indicators to sales outcomes. AI analysis of the data has allowed Mars to develop an 85% level of accuracy in predicting whether a new creative will drive sales results.

Elsewhere, BMW Canada has deployed carbon neutral programmatic advertising that uses machine learning to optimise campaigns not just for performance but also to lower carbon dioxide emissions and directly offset emissions approach. The truth is that AI will significantly impact both the field of marketing and far broader levels of business operation.

All organisations need to be thinking about how to incorporate AI in their day-to-day activities. But as it stands, too much of the discourse is spent speculating as to whether or not AI will take existing jobs (it will) and there not enough focus is on what jobs it will create.

We’ve been here before. The advent of the Internet, and Web 2.0 after it, changed the way we work and made some jobs irrelevant. The Industrial Revolution did the same. AI is new but technology disrupting the workforce is not. As children, we all heard the stat that 85% of the jobs we will do don’t even exist yet. It’s a very approximate figure, but the principle hasn’t changed.

Understanding that this latest change will bring a critical skills gap is where marketers must start. This doesn’t necessarily mean restaffing – in this market resourcing budget is hard to come by – but it does mean opportunity for those marketers who find ways to institutionalise curiosity around AI in their teams, encouraging and empowering their staff to learn how to use the tools and understand their benefits and limitations.

If you’re yet to begin your journey, the best place to start overcoming barriers is to simply start small. It’s not just about understanding what off-the-shelf tools are actually available, and discovering how they may optimise certain functions. A critical challenge to using AI tools effectively is being able to actually articulate what it is you want to do.

We’re all aware of roles like Prompt Engineer hitting the job market for this reason, but these skills are like muscles that can be developed across a whole team with practise and the freedom to experiment. Running innovation sprints or carving out semi regular time for the team to down tools on their regular work and brainstorm small proofs of concept are fantastic initiatives.

If you’re past the “start small” stage, new challenges will arise, largely concerning operational implementation. If you’re finding yourself struggling to prioritise use cases amid competing demands on time and resources within a large business, you’re not alone. Teams often have backlogs of initiatives vying for development slots. Determining which AI applications provide the most value therefore becomes crucial.

One approach is start simply by focusing on proofs of concept that address clear, narrowly defined problems, or focus on optimising a specific metric like conversions or loyalty. This is the shortest path to demonstrating real benefits, measuring success and securing further investment from the business.

From there, collaboration across functions – like data-focused monthly meetings to share progress with the rest of the business – become a critical factor to scaling up the role of AI. Bringing together different perspectives helps unearth the most impactful cross-departmental uses of AI, and if the results are clearly defined, they’re difficult to ignore.

Applying a rapidly emerging technology that demands adaptability and, quite frankly, a significant amount of time is a daunting job, particularly for already time-poor CMOs.

But starting with experimentation and curiosity will yield startling results, particularly if initiatives remain data-focussed and set a clear idea of what success looks like early on. By building the talent and organisational muscle now, marketers are setting their businesses up to reap serious rewards in the not so distant future.

So, whether you are an AI novice in awe of the possibilities or have been working with AI for some time and recognise its limitations, everyone needs to be thinking about how AI can be continually leveraged in their business, in marketing and beyond. We are nearing the peak of the AI hype cycle, which means the current hype and most of the AI technology we are all excited about today is not going to last the test of time.

But it’s undeniable that AI, in some form or another, is here to stay and those who embrace it and learn to harness it will succeed.

Article originally published on Mumbrella.