“AI readiness” tends to be used as umbrella phrase to describe the different range of activities that businesses need to carry out to get value of generative AI and minimize any associated risks.
AI readiness therefore takes into account everything from having the right data foundations and governance in place to supporting AI literacy among staff to establishing ethical guardrails.
One area where businesses – including even large brands – are collectively not ready for AI is ensuring their brand is visible and faithfully represented across Large Language Models (LLMs) and in responses from associated AI tools like ChatGPT, Gemini and Claude.
The outside-in view
Too many large companies have a digital estate that contains content that is outdated or contradictory. Because AI tools base their responses on the content they find, brands risk AI tools giving out misinformation or misrepresenting their brand, products or organization.
Similarly, there is also a risk of brands not being featured in responses or less than those compared to competitors, impacting future visibility and competitiveness.
This is effectively AI looking from the “outside in” and then presenting a picture of what it has surfaced. When we measured this “outside in” element of AI readiness across the websites of over 250 top global brands using an automated remote website assessment tool and an underlying AI maturity model, we found only approximately 3% of websites could be considered “leading” when it comes to AI readiness.
Why AI readiness is important
The way people search for information on the web is changing rapidly. Consumers are turning to ChatGPT rather than using traditional Google searches; surveys have suggested that 37% of consumers in the US who already use AI, now start their searches using AI tools rather than Google.
This means that rather than looking at a web page on a company’s own website, consumers’ perceptions about a brand are increasingly coming from AI responses as the first step rather than directly from a web page controlled by the company.
That’s a problem if a consumer asks AI to provide a list of suppliers or vendors, and your business isn’t listed when it should be, or the AI provides erroneous information about your company and products from the off.
Recognizing the speed at which consumers are turning to AI tools for search, digital marketing teams who have previously invested in Search Engine Optimization (SEO) are now investing in related practices like Generative Engine Optimization (GEO) and Answers Engine Optimization (AEO) with some urgency to optimize their digital footprint so it is successfully reflected in AI tools like ChatGPT.
But the reasons that even some of the world’s top brands still have a lot of work to do in this area is because in the past they have deprioritized basic governance and maintenance across their digital estate, with the risk of AI confidently presenting answers as facts or broadcasting perceptions that are based on information from content that is out of date and has been superseded.
Of course, the risks associated with erroneous answers may vary but they can be acute, particularly in heavily regulated industries such as financial services, healthcare or pharmaceuticals with strict compliance surrounding the information that businesses must provide to customers, or in areas such as products where health and safety instructions must be correct.
The hidden digital footprint
One of the key reasons where businesses slip up is by building up a digital footprint – often without realizing it – which is not actively managed and has old and inaccurate information.
Over time, businesses accumulate a surprisingly large number of digital assets that are accessible online, sometimes hosted on third-party platforms: websites for sub-brands or locations, online newsletters, campaigns, product launches, events, microsites, partner content, customer portals, forms, digital experiments, and so forth. In addition, there might be areas of a corporate website that have effectively been forgotten about.
Over the years, this digital footprint in some organizations has not been properly managed, so a site created to support a one-off event stays online indefinitely rather than being retired. Sometimes when a company acquires another, it also becomes responsible for this digital debris that has accumulated and which the new team have little awareness of the full footprint they have inherited.
Previous research we carried out between 2017 and 2023 shows that central digital teams can be unaware of up to 41% of their entire digital footprint.
The danger of poor document management
Another critical cause of a lack of AI readiness is an overreliance on documents. Many websites contain PDF attachments which don’t necessarily get updated and then accumulate over years and years, containing information which is no longer relevant or correct.
Annual reports, brochures, manual forms, product guides and more are all left accessible on websites. Our research suggests that as many as one in five (19%) website documents are duplicates which can create inconsistent signals for AI about which is the current and authoritative version
Not enough organizations take a structured approach to managing their documents across their digital estate, which means they cannot effectively update documentation for example to reflect brand, regulatory or operational changes, with AI now forming responses from these out-of-date files.
How to reduce risk exposure
As more businesses find AI is producing misinformation about them and the number of consumers who use AI tools for searching information increases, awareness of this problem is likely to rise.
To reduce the risk exposure, at a high-level, three things generally need to happen:
- The issue needs to be owned at the executive or board level with someone responsible to tackle it.
- Digital teams need to use automated assessment tools to measure and monitor the risk associated with their entire digital estate in order to track associated remedial action.
- The digital team need to put governance and lifecycle management processes in place that enables them to have more control and oversight over their digital footprint, so it avoids becoming a digital wild west again.
AI readiness covers many areas. A key priority for brands who want to maximize AI visibility while reducing the chance of misinformation is to get their digital estate under control. Of course, there are other important activities to consider including ensuring content is optimally structured, but reducing the amount of out-of-date content is a critical starting point.
We’ve featured the best AI chatbot for business.
This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.
The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit
