The adoption of AI has become synonymous with progress and innovation. Just over half (59%) of companies surveyed in IBM’s 2024 Global AI Adoption Index are exploring or deploying AI and plan to accelerate their investments in such projects. But, in their eagerness to adopt this new technology, some businesses’ AI projects are ending in failure.
McDonald’s recently decided to abandon its AI drive-through trials after videos of mistaken food orders went viral, including a customer receiving ice cream topped with bacon and another receiving £166 worth of unwanted chicken nuggets. The fast food giant, which has been testing voice recognition technology to process orders since 2021, will be removing its AI system by the end of August.
In February, Air Canada had to compensate a customer after its chatbot misinformed him about its refund policy. A small-claims court agreed with his claim that the AI assistant had falsely assured him that he could secure a discounted air fare.
High-profile mishaps such as these are costly and can tarnish a company’s reputation. For some, they signal the need for a more measured approach to AI in order to mitigate risks and help to build a more sustainable strategy that evolves with the business needs. Others, meanwhile, would argue that businesses cannot afford to sit idly on the side lines and should accept failure as an important and sometimes unavoidable part of the experimentation process.
Slow and steady wins the race
In the haste to adopt AI, organisations often grapple with increased hardware requirements and ballooning cloud costs. The need for more capacity and speed can escalate quickly, leading to higher usage costs.
But the financial strain is just one side of the coin. A hurried approach to AI integration can result in misaligned use cases, unclear objectives and poorly-defined timelines. These missteps can derail even the most promising AI projects from the start.
Rushing to the finish line without adequately preparing end-users can also hinder adoption success. Without clear communication of value and robust training, users may struggle to leverage the AI solution effectively, leading to under utilisation and potential failure of the initiative.
Rushing to the finish line without adequately preparing end-users can hinder adoption success
In the race to integrate AI, slow and steady is a strategy that allows organisations to scale their tech infrastructure at a sustainable pace, effectively managing costs and avoiding unnecessary upgrades.
Choosing the right use cases from the start requires careful thought and planning. A clear objective, return on investment and timeline can only be established with a measured approach. Rushing this process can lead to misaligned projects that veer off course. The finish line of AI integration is just as important as the starting point. A measured approach provides ample time to articulate the value proposition to end-users, ensuring they receive comprehensive training and consistent reinforcement. This paves the way for successful adoption and utilisation of the AI solution.
Taking a measured approach to AI isn’t just smart, it’s strategic. This method will likely result in improved operational efficiency and strategic growth and also ensures a perfect blend of old and new skills within IT teams, keeping companies ahead of customer needs and increasing loyalty.
Beyond these immediate benefits, a paced approach to AI integration fosters a culture of continuous learning and adaptation. As AI evolves, so must our understanding and application of it. By not rushing into an AI-driven future, we create space for thought leadership, innovative problem-solving and the development of robust strategies that stand the test of time. That’s the real power of patience in AI integration.
Businesses are not moving fast enough
AI isn’t just the buzzword for the year, it has shown real, tangible benefits including productivity, growth and innovation. But there’s no reward without risk.
Businesses are not moving fast enough to implement this technology, which means they’re losing out on the benefits and risk being outpaced by other companies. Furthermore, In a tight labour market, ignoring the AI hype can hamper a businesses ability to recruit and retain talent.
In instances where AI can be used to improve customer service, it can also be used by bad actors. This can be seen with the rising use of deep fakes. For sectors such as banking, where there’s been scaremongering around fraud and AI, the only way to fight increasingly sophisticated cyber crime is by leveraging those very same AI tools and technologies.
The benefits this technology can deliver are too lucrative for us to ignore
But by its very nature, AI is designed to learn and improve. The more people are using AI, the smarter the tech gets. It is important that businesses begin experimenting, prompting and better understanding this innovative technology now. The more we can leverage it, the smarter it will get and greater the impact it can have.
While I would encourage businesses to embrace AI, that does not mean business leaders should forget to safeguard their workforce and company. This includes creating the right culture. Providing open communication and upskilling employees is important to helping employees use AI effectively.
Businesses must recognise the importance of putting AI at the heart of their transformation strategy, not merely using it as a ‘proof of concept’ to address individual challenges.
The reality is that AI will be used for both good and bad. The negative consequences are likely to happen whether businesses adopt AI or not. The benefits this technology can deliver are too lucrative for us to ignore.
