Emerging Artificial Intelligence Technologies of 2020s: Formative AI

Artificial Intelligence Artificial Intelligence Artificial intelligence technologies have always lived up to the hype. So what’s next?

The innovation in technology has been a direct response to demands and expectations from humans. Emerging technologies like artificial intelligence continues to play a vital role in helping the world in numerous ways. Even artificial intelligence is evolving through many iterations. Recently, Gartner has considered ‘Formative AI’ in its Hype Cycle for Emerging Technologies 2020. So, what is Formative AI? Let’s learn more about it.

What’s Formative AI?

Gartner defines ‘Formative AI’ as a type of artificial intelligence that is capable of dynamically changing itself to respond to a situation. In layman’s terms, it refers to artificial intelligence technologies which dynamically adapts based on context. There are a variety of types, ranging from artificial intelligence that can dynamically adapt over time to technologies that can generate novel models to solve specific problems. So basically, apart from carrying complex analysis, Formative AI has potential to create things by itself – adding our fears of artificial intelligence replacing humans.

The three most common forms of Formative AI include: Generative AI, AI-assisted software development and Generative Adversarial Networks (GANs). GANs can also be considered as an application of Generative AI. According to Gartner, other emerging technologies in this trend include composite AI, differential privacy, small data and self-supervising learning. These technologies can be used by even enterprises looking to experiment with artificial intelligence and can leverage AI-augmented development, AI-assisted design, adaptive machine learning, ontologies and graphs, and even generative AI. This indicates that the applications of Formative AI range from synthetic data generation, drug discovery to train machine learning models or even design deepfakes with despicable intention. Gartner also predicts Formative AI will be used to streamline how mathematical and machine learning models are created and fine-tuned over time.  Gartner’s Hype Cycle for Emerging Technologies explores the opportunities for organizations in their search for technology-enabled business transformation.

Hype and Concerns Behind Formative AI

It is true that implementing artificial intelligence technologies will warrant significant investment for companies, but the rewards can far outweigh the initial outlay. While older applications of artificial intelligence like machine learning, computer vision, offer benefits beyond business profit, Formative AI offers freedom to major extent. For instance, suppose researchers need dataset to train an AI model – that is not available readily or is insufficient, then they can employ Generative AI to create needed data for good predictions or analysis. Even AI augmented development can empower companies to mitigate staffing and training concerns that hinder AI adoption or scaling.

Moreover, Formative AI helps to reduce the costs of machine learning for enterprises, making it extremely attractive to cost-conscious businesses.

One of the popular examples of Generative AI is OpenAI’s GPT-3. GPT-3 can create slide presentations, images; write programming code, articles and parse and translate texts and much more. There are nefarious uses of Generative AI through creation of deepfakes that causes security and privacy concerns. Hence, DARPA is pursuing the detection of deepfakes to counteract fraud, disinformation, instigation of social unrest.

Typically, most of the Gartner’s predictions are proven true with the passage of time. So, if the pursuit of Formative AI goes well, it will have a huge influence on the trajectory of applications of artificial intelligence in future. However, to achieve this we need to pay heeds to another Gartner’s Hype cycle technology i.e. Algorithmic Trust. As Gartner puts it “Models based on responsible authorities are being replaced by algorithmic trust models to ensure privacy and security of data, source of assets and identity of individuals and things.” So, we need to have trust algorithms in place to trust the artificial intelligence algorithms for the save use of Formative AI. Else the ‘hype’ may fizzle out soon!