Two key problems for businesses hoping to reap the benefits of generative AI have remained the same over the last 12 months: hallucinations and trust.
Business leaders need to build trustworthy applications in order to harvest the benefits of generative AI, which include gains in productivity and new ways to deliver customer service. To build trustworthy AI applications that don’t ‘hallucinate’ and offer inaccurate answers, it helps to look at internet search engines.
Internet search engines can offer important lessons in terms of what they currently do well, like sifting through vast amounts of data to find ‘good’ results, but also areas in which they struggle to deliver, such as letting less trustworthy sources appear ahead of reliable websites. Business leaders have complex requirements when it comes to the accuracy needed from generative AI.
For instance, if an organisation is building an AI application which positions adverts on a web page, the occasional error isn’t too much of a problem. But if the AI is powering a chatbot which answers questions from a customer on the loan amount they are eligible to, for example, the chatbot must always get it right otherwise there could be damaging consequences.
By learning from the successful aspects of search, business leaders can build new approaches for gen AI, empowering them to untangle trust issues, and reap the benefits of the technology in everything from customer service to content creation.
Finding answers
One area where search engines perform well is sifting through large volumes of information and identifying the highest-quality sources. For example, by looking at the number and quality of links to a web page, search engines return the web pages that are most likely to be trustworthy.
Search engines also favour domains that they know to be trustworthy, such as government websites, or established news sources.
In business, generative AI apps can emulate these ranking techniques to return reliable results.
They should favour the sources of company data that people access, search, and share most frequently. And they should strongly favour sources that are known to be trustworthy, such as corporate training manuals or a human resources database, while deprioritising less reliable sources.
Building trust
Many foundational large language models (LLMs) have been trained on the wider Internet, which as we all know contains both reliable and unreliable information.
This means that they’re able to address questions on a wide variety of topics, but they have yet to develop the more mature, sophisticated ranking methods that search engines use to refine their results. That’s one reason why many reputable LLMs can hallucinate and provide incorrect answers.
One of the learnings here is that developers should think of LLMs as a language interlocutor, rather than a source of truth. In other words, LLMs are strong at understanding language and formulating responses, but they should not be used as a canonical source of knowledge.
To address this problem, many businesses train their LLMs on their own corporate data and on vetted third-party data sets, minimising the presence of bad data. By adopting the ranking techniques of search engines and favouring high-quality data sources, AI-powered applications for businesses become far more reliable.
A swift answer
Search has become quite accomplished at understanding context to resolve ambiguous queries. For example, a search term like “swift” can have multiple meanings – the author, the programming language, the banking system, the pop sensation, and so on. Search engines look at factors like geographic location and other terms in the search query to determine the user’s intent and provide the most relevant answer.
However, when a search engine can’t provide the right answer, because it lacks sufficient context or a page with the answer doesn’t exist, it will try to do so anyway. For example, if you ask a search engine, “What will the economy be like 100 years from now?” there may be no reliable answer available. But search engines are based on a philosophy that they should provide an answer in almost all cases, even if they lack a high degree of confidence.
This is unacceptable for many business use cases, and so generative AI applications need a layer between the search, or prompt, interface and the LLM that studies the possible contexts and determines if it can provide an accurate answer or not.
If this layer finds that it cannot provide the answer with a high degree of confidence, it needs to disclose this to the user. This greatly reduces the likelihood of a wrong answer, helps to build trust with the user, and can provide them with an option to provide additional context so that the gen AI app can produce a confident result.
Be open about your sources
Explainability is another weak area for search engines, but one that generative AI apps must employ to build greater trust.
Just as secondary school teachers tell their students to show their work and cite sources, generative AI applications must do the same. By disclosing the sources of information, users can see where information came from and why they should trust it.
Some of the public LLMs have started to provide this transparency and it should be a foundational element of generative AI-powered tools used in business.
A more trustworthy approach
The benefits of generative AI are real and measurable, but so too are the challenges of creating AI applications which make few or no mistakes. The correct ethos is to approach AI tools with open eyes.
All of us have learned from the internet to have a healthy scepticism when it comes to facts and sources. We should be levelling the same level of scepticism at AI and the companies pushing for its adoption. This involves always demanding transparency from AI applications where possible, seeking explainability at every stage of development, and remaining vigilant to the ever-present risk of bias creeping in.
Building trustworthy AI applications this way could transform the world of business and the way we work. But reliability cannot be an afterthought if we want AI applications which can deliver on this promise. By taking the knowledge gleaned from search and adding new techniques, business leaders can find their way to generative AI apps which truly deliver on the potential of the technology.
- Data & AI
