Harnessing the power of AI to combat fake reviews [Q&A]

There's a lot of talk about how artificial intelligence (AI) is changing the world. And it's true -- AI has already started transforming the healthcare, finance, and manufacturing industries. But there's one area where AI is causing some severe problems: fake reviews.

We spoke with Truely CEO JP Bisson about AI technology and how companies can use it to protect their interests.

BN: How has AI technology impacted the way we live and work?

JB: Artificial intelligence had come a long way since the 50s, when Alan Turing asked, 'Can machines think?' Although no machine has even come close to passing the Turing Test, that question sparked the decades-long research that has brought us AI technology as we know it today.

We are probably unaware of it, but AI is now a considerable part of our everyday lives. We can play a song, turn house lights on and get answers to questions using voice assistants. We get personalized content on our social media feeds and video streaming services. We unlock our phones and open our apps with facial recognition. We get to our destination in self-driven vehicles.

At work, we have software programs that automate operational processes, allowing companies to save resources. We have tools that analyze data, software that measures employee experience, robots that speed up product manufacturing, apps that write content. Recently I have heard of two AI systems in the works -- one that produces realistic visuals and artwork from a description in plain language and another that automates manual tasks in-browser.

It is all so amazing that I can't wait to see what happens next. Still, we must look at AI as a double-edged sword that could cause grave harm if it falls into the wrong hands.

BN: In what ways can AI cause this harm?

JB: We live at an exciting time when almost everything we set our mind to becomes possible, especially with AI technology. But just like I said, people can also use artificial intelligence to harm and destroy the things the technology has helped build.

Case in point: deep learning AI is now used to create deceptive images and videos or deepfakes. AI writing tools have been taught to generate false news and fake reviews. What's scary about all these fakes is that they seem real. To the unsuspecting reader or viewer, they are believably true.

The internet has become a great source of information, and rightfully so. You only need to type a word on any browser, and you'll get pages of results. So you can't blame why people who aren't so discerning believe almost everything they read and see on the web. Unfortunately, the internet has developed into a medium where people no longer necessarily tell the truth.

These untruths have led us to develop a site that would help bring truth back to the internet. We understand how fake reviews hurt not just the product and service providers but also the consumers. Truely's aim is to stop these fake reviews -- which can be mass produced by AI -- and bring the truth back to the internet. And we are fighting fire with fire, using AI technology to identify and remove these fake reviews -- whether written by paid humans or AI-generated -- to give consumers the real score about the products and services we feature on our site.

BN: Can you tell us more about the AI technologies used to create fake reviews, how these technologies work, and how artificial intelligence can filter out fakes?

JB: Generally, artificial intelligence has four components: machine learning (ML), natural language processing (NLP), computer vision, and robotics. Out of those four, people use ML and NLP to generate text.

Machine learning allows computer computers to learn from data without being explicitly programmed. This AI system is based on the idea that machines should be able to learn and improve independently, just like humans.

There are two main types of machine learning:

• Supervised learning, where the computer is given a set of training data. The computer then uses this training data to learn what it's supposed to do.
• Unsupervised learning, where the computer is given data but is not told what to do with the data. It has to figure out patterns and structure on its own.

Under machine learning, we have deep learning, which is a powerful tool for generating text. So you feed a computer a large body of text, and from there, the system learns the distribution of words and grammar rules.

Once the system has learned the rules and what-have-you, it can generate new realistic and coherent text. And yes, this technology has been very helpful in quickly creating written content such as news articles, product descriptions, text summaries, and more. Unfortunately, people also use it to create fake news, reviews, and comments -- which, as we know, can be detrimental not just for businesses who get these fake ratings but for society as a whole.

Natural language processing, on the other hand, is a form of artificial intelligence that allows computers to understand and process human language. This system works by converting human language into a format that computers can understand. AI algorithms are then used to analyze and interpret human language. With these algorithms, a computer can understand the overall meaning of a text, identify named entities, extract key phrases, and even find the sentiment of a text. NLP is becoming more and more commonplace in our everyday lives, with people using it for the same use cases they use ML for.

At Truely, we use these technologies to filter out fake reviews. Our AI analyzes reviews for certain red flags and language patterns, indicating that the reviews are not genuine. It also looks for patterns in the language used. For example, fake reviews often use superlative words to describe a product, while real reviews are more likely to use neutral language. It also uses other clues to identify and filter out fake reviews on the list of software products and services on the Truely site.

The tools we use have proven to be valuable in the fight against fake reviews. As AI technology evolves, I am optimistic that it will become more reliable in weeding out fake reviews.

BN: What does this technology mean for SaaS providers and businesses?

JB: There is only one thing I'm sure of: the world of online reviews is changing. This AI technology is already significantly impacting businesses, especially those that rely heavily on online reviews.

Generally, this AI technology is good news for businesses, as it means they no longer have to worry about the negative impact of fake reviews.

Companies that rely on fake positive reviews to boost their ratings will now be more likely to be exposed, leading to a loss of customer trust and potential legal problems.

Overall, AI that removes fake reviews is a significant development for businesses. Besides exposing fake reviews, it protects the credibility of SaaS firms and other companies.

Photo Credit: Bacho/Shutterstock