The European Union’s (EU) landmark AI Act, recently passed by the European Parliament, is described as the first ever comprehensive legal framework on Artificial Intelligence (AI) worldwide (Brazil might argue this claim, though none of the four bills proposing a framework for the ethical and responsible use of AI systems have been approved by Congress at the time of writing).
It has significant and wide-ranging implications for the tech industry, businesses, and consumers, and guides what we should expect from future customer experiences.
The EU AI Act: A brief overview
The EU AI Act aims to regulate artificial intelligence systems by prohibiting AI applications that carry unacceptable risks to safety, livelihoods, and the rights of EU citizens.
This may seem entirely sensible and plays into the many debates about using AI, machine learning, deep learning, and neural networks in obvious areas such as biometric identification, “deep fakes”, and behavioral manipulation. However, how the Act is written places additional obligations on all businesses that plan on using AI in many less controversial ways – including in the generation of customer experiences and customer communications.
A powerful precedent
As with other recent EU-wide rules such as GDPR, the AI Act casts a wide net. Professional organizations based in the EU, or with any customers in the EU, will need to comply with the regulations. With large global organizations generally being risk-averse, the knock-on effect is likely to be that EU-wide laws will become generally accepted global practices.
Articles 2 and 52 allow for some exceptions to the rules, notably in government and international security, but most industries that plan on having any sort of European presence will need to be fully compliant.
A sea-change in future tech direction?
Despite high-profile incidents where AI has failed, including situations where history has been subtly rewritten, where information with racist overtones has been given, or where information provided was completely incorrect, organizations have marched towards an AI-generated future under the pretext that “the computer knows best” and that customers can trust the advice given. The new Act challenges this perception and, among other areas, requires that enterprises communicate their commitment to compliance and quality, and verify the accuracy and legality of all automatically generated content.
Crucially, the catchily titled “Title IV, Article 52: Transparency Obligations for Providers and Users of Certain AI Systems and GPAI Models” prohibits organizations from using most AI-generated content (including in written form) without it being clearly labeled as such.
This means that the envisioned future of self-generating communications fed from a powerful data-rich AI hub is still technically possible, however, the customer experience is likely to be called into question. We have all been raised to know that “it’s the thought that counts”, and if full marketing and servicing communications are obviously generated without any human touch being added to the content creation, the longer-term relationship with the customer is likely to suffer.
Working within the new rules requires a small but important change: automation needs to become augmentation. Communication and experience designers should use AI to generate personalized, readable, and meaningful content quickly, and then let the human experts review the content and place it into templates for delivery across all digital and physical channels. This process offers the best of both worlds – an ability to rapidly create and amend content for all communication types, reducing the overheads traditionally associated with customer communication management, and a promise of accuracy and authenticity due to human oversight of the process. It also conveniently removes the need to label any communications as being AI-generated.
From a brand perspective, human oversight ensures that your communications continue to honor the promises of your brand’s tone and style, with the opportunity to address any AI-generated output that feels too generic and detached or doesn’t align. This ensures that the important nuances of brand identity and messaging objectives are reflected accurately, in a way that technology alone cannot always achieve.
Start your journey into AI augmentation
The EU AI Act disrupts traditional ideas about the future of AI-generated interactions but offers new opportunities for AI-augmented human-driven processes. Embracing these changes early will allow businesses of all sizes to navigate the new regulations while transforming their communication-related experiences and maintaining positive customer relationships.
Inspire Evolve leverages AI to empower you to deliver customer communications with impact. By providing sentiment and tone-of-voice analysis to ensure your communications are on brand and accessible, as well as operational benefits such as the ability to translate and repurpose content for various applications, your customer communications could not be in better hands. There's a reason why we're ranked the world's leading CCM provider!
Speak to a Quadient expert today to find out more about how Inspire Evolve can help you.
Finally, for the avoidance of doubt… this blog was not written by AI!
