Skip to main content


applied intelligence Live

07 Aug 2023

Safeguarding AI: Building Ethical ChatGPT | Key Risks & Solutions

In partnership with Omdia and AI Business
Safeguarding AI: Building Ethical ChatGPT | Key Risks & Solutions

Explore AI's ethical dilemmas with ChatGPT. Get insights on reducing bias, understanding regulations, and ensuring fairness in generative AI tech.

Generative AI accounted for 1% of all data in October 2021. By 2025, that figure could rise to 10%, and by 2027, 30%. The market for generative AI is predicted to be worth trillions in the next decade. A new report from Informa Tech's Applied Intelligence Group ‘Beyond ChatGPT: Building Responsible Generative AI for Business’ highlights generative AI’s impact on creativity, copywriting and data augmentation.

Identifying Key Risks of ChatGPT Use

While generative AI could augment the way we work, it comes with potential risks. The report addresses the following issues:

Implementing Fairness and Transparency Protocols

Generative AI could mean the automation of some roles risking job losses. But while generative AI can lead to job displacement in some areas, it can also create new job opportunities. For example, it has triggered a growing need for skilled professionals in AI like machine learning specialists and prompt engineers.


While transformative, generative AI is at risk of perpetuating human biases.

AI models are only as good as what they are trained on, meaning historically biased data input into the system could be used to perpetuate prejudices. The large majority of AI developers are male but there’s also the risk of biases creeping in related to race and language.

Strengthening Privacy Policies and Data Protections

A tool as easily accessible as ChatGPT could be used to spread misinformation. We’re already starting to see this issue arise from something as simple as a lack of fact-checking to purposefully created fake social media posts, the harm this could cause is catastrophic, especially considering an election cycle is approaching.

Intellectual property issues

Generative AI works by essentially copying existing work. What you get out is a reformation of what you put in. There are concerns that large numbers of models were trained on web-crawled data without permission from authors and copyright holders.

These are just some of the top level insights from the report. Get full access here.

Want to learn more and take part in the discussion on Generative AI? You can join us at this year's Applied Intelligence Live event (20-21 September). Check out sessions on ‘The Future of Enterprise Creativity: How Generative AI is Changing the Game’ and explore how businesses can harness AI while managing the risks that come with it.


Explore Our Full Blog Library