Articles > AI for Business - Your 5 reads of the week #4

3 months ago


Hello readers and welcome back to our weekly AI for Business reads.

It is already our 4th edition, time flies when you are doing what you love. In this edition, you will learn about different expectations experts have on AI, how the EU intends to win the AI race thanks to ethics, why communication between business people and engineers fail, how AI can stop money laundering, and an AI describing the environment for visually impaired people.

Enjoy!

FIGURE OF THE WEEK

While only 33% of consumers think they are using AI, the study shows that 77% of devices we use today have features that use some forms of AI. This clearly shows that there is still a major gap between consumers awareness of AI and its actual application in daily life. Showing this number can help your customers understand that AI has been an integrated part of their lives and should not be mistaken as a hostile innovation that is only there to take their jobs.

AI WILL CHANGE BUSINESSES

'Should we be afraid of AI?' or 'Can we trust AI with our data?'. These questions continue to arise and the EU decided to write a checklist in order to have a general guideline of what is and isn't possible when talking about AI. This is merely a guideline and has no juridical power, but it is a step in the right direction for European companies who want to know if their AI model is considered ethical. These guidelines were written by an independent group of 52 experts and they are now inviting companies to check if their AI model complies to the list.

Key requirements are: transparency (why the AI model recommends or takes certain decisions), robustness and safety (AI systems shouldn't be vulnerable to hostile attacks), and discriminatory actions (no gender or race bias).

Every new technology goes through 4 stages: inclination, early adoption, wide acceptance, and lastly, either the technology declines, or new application areas are found, and so the cycle repeats itself. alt text

Likewise, the same applies to Artificial Intelligence. We are now in the decline phase because there is a major issue on the interaction between Machine Learning programmers and Business Developers. Chief strategy officers or business developers have aspirations and anecdotes phrased in an obscure language that is seen by ML engineers as “corporatese”. This gap needs to be filled for the development of new and practical AI applications. This article proposes that companies change how they communicate and how they frame problems. Business profiles should be trained to understand computational and algorithmic concepts. This means being able to convert business problems into “computational business problems”, which are understandable for technical profiles. Vice versa, technical profiles like engineers should also further develop their soft skills and business mindset. There is an increasing need for bilateral efforts to bridge the expertise of business and technology/science. The article ends with a perfect quote: “Do you have an AI strategy?” makes as much sense as asking, “Do we have an Excel strategy?” Business people within companies need to recognise the importance of being more familiar with technical terms.

By now, we all know the positive impacts AI can have in different fields, and in different aspects of business, but there exists also criticism about AI. One is raised by Kai-Fu Lee, a world-renowned Taiwanese venture capitalist and expert on AI, who believes s that in 15 years, 40% of jobs could be displaced by AI. Many experts have countered this argument by pointing out that these jobs are dull and repetitive ones, and AI, taking over them, will allow more time for creative work, which will also be more fulfilling for the employees. Another side of doing business is that you need to stay competitive and relevant in the market, and so, you need to differentiate yourself. One competitive advantage is by lowering costs and it can be achieved by embracing automation, which causes labour workers to be worried. Like many digital transformations, AI is going to change the way businesses look at automation. Our opinion at Kantify is and always will be: AI should be seen as a tool for humans to make better decisions.

AI IN PRACTICES

Money laundering is a major issue all over the world. The United Nations Office on Drugs and Crime reports that it accounts for up to 5% of global GDP. Countries are searching for ways to combat money laundering because a lot of the times it is linked to the funding of human trafficking, or large criminal organisations. One way criminals launder money is by hacking ATMs and using money mules (local criminals) to pick up the money ATM releases.

These small transactions could be (willingly) overlooked by humans, and thus, are not reported. This is where AI could step in and remove the possibility of bribery out of the equation, by objectively detecting suspicious activities on a large scale. AI could also spot fake accounts which are set up remotely by bots.

An app that describes a picture for visually impaired people has been built by Google. Once again, Artificial Intelligence has shown that it can make the world a better place. We have already read cases about AI's helping radiologists be more human or improve the Mcdonald's menu screens, but now Google has taken the initiative to help the less fortunate: the visually impaired people. They have built an app to describe whatever the camera sees. This gives the users the ability to walk around with their dog and have an idea of their surroundings.

And do not forget to subscribe to our newsletter for your monthly AI dose.

See you next week!

alt text