Sannidhya Agrawal
3 min readJan 11, 2023

Generative AI: From Data Generation to Creative Intelligence

A common idea that our creativity is what makes us uniquely human has shaped society but strides of progress made in the domain of Generative Artificial Intelligence question this very notion. Generative AI is an emerging field that involves the creation of original content or data using machine learning algorithms.

As we think about a future where humans and AI partner in iterative creative cycles, we consider how generative AI could impact current businesses and possibly create new ones. Up until recently, machines were relegated to analysis and cognitive roles, but today algorithms are improving at generating original content. These technologies are iterative in principle, one is built on top of the last one, and each new iteration enhances the algorithm and increases the potential for discovery exponentially.

The technology presents itself as a more refined and mature breed of AI that has sent investors into a frenzy and among all this emerges a clear market leader — OpenAI. Its flagship products- ChatGPT and DALL-E proved to be industry disruptors and brought generative AI tools to the masses. DALL-E allows people to generate and edit photo-realistic images simply by describing what they want to see, while ChatGPT does the same through a text medium.

Images generated by DALL-E -a deep learning models developed by OpenAI to generate digital images from natural language descriptions

This type of A.I. holds the potential to reinvent almost everything and facilitate a new way of interacting with an array of software. It opens new avenues for creating an entirely AI-based hyper realistic consumer experience. Text is the most developed domain. However, producing content natural language is difficult to achieve and quality is often hampered. Code generation is likely to improve developer productivity in the future but full stack software development remains a distant dream. Advancements in areas of 3D modelling could very well equip generative AI with tools to work in physical product design and eventually into full-fledged Game Engines. There are great strides being made in many fields, from audio and music to biochemistry.

The dream would be that generative AI brings the marginal cost of creation down to virtually zero, greatly improving the productivity of the workforce and adding economic value, but this raises a greater ethical dilemma. A grey area is whether generative A.I. may affect copyrights and whether the companies need to get permission to use the data that trains their algorithms.

Tech giants like Google have been reluctant to release their generative technologies as widely as OpenAI to shield their established corporate brand image as these systems tend to produce toxic and malicious content that leads to social disharmony. The lack of content moderation mechanisms and safeguards can harm and adversely impact society through misuse, perpetuating biases, exclusion, and discrimination. A biased technology hampers the very idea of using generative AI independently in the domain of entertainment and journalism.

For example, initially, generative imagery would show only images of people of colour for prompts like “prisoner or criminal.” Generative AI systems can create content for malicious purposes, such as deepfakes, disinformation, and propaganda. These can be used to polarize the masses, swing political ideologies, and spread misinformation. Even the most minute errors and inefficiencies by generative AI in the field of biochemical research can lead to cataclysmic consequences. These technologies have technical limitations as well, they are large and difficult to run (requiring high computational power and GPU orchestration), not broadly accessible (unavailable or closed beta only), and expensive to use as a cloud service.

A more transparent model needs to be promoted about the working of AI algorithms and higher accountability of the firms that produce these technologies is the need of the hour. Systems should be developed with human oversight to promote collaboration between AI and humans. AI today may not be able to solve real world problems like creating efficient designs and optimized codes. Will generative AI limit human imagination or stop the cycle of innovation? A lot of questions remain unanswered as we move forward into the future with a cautious approach.