Features Hub Comment

AI Ethics in Business: Strategies for Responsible AI Use

Wed 31 May 2023

Just a decade ago, the rapid impact of artificial intelligence (AI) and the crucial importance of AI ethics in today’s business world were beyond even the most optimistic predictions of AI experts.

Innovations ranging from the launch ChatGPT in November 2022, to automation solutions implemented across businesses of varying scales, have shaped a rapidly developing technological landscape.

In this swiftly evolving ecosystem, AI applications are subject to minimal regulation, and businesses often shoulder the responsibility of defining their AI utilisation protocols. This has raised concerns about how unregulated use of AI tools can yield discriminatory outcomes, compromising precision and accuracy.

Prominent Voices in AI Ethics

Dr Lynn Parker, a leading expert in AI who served in the White House Office of Science and Technology Policy between 2018 and 2022, believes that considerations about trust and ethics in the development of AI technologies have come too late in the process. She said that ethics was rarely discussed within the AI research community at technology conferences until five years ago.

Geoffrey Hinton, dubbed the ‘godfather’ of AI, also resigned from his job at Google, saying that AI technologies could be used to harm people and spell the end of humanity.

A study from Pega showed that just 12% of respondents believe AI knows the difference between good and evil. More than half of consumers also do not believe that AI can behave morally (56%) or make unbiased decisions (53%).

With the growing emphasis on AI ethics and regulations, the time is now for businesses to start thinking about how they will use AI responsibly.

Business Implications of AI Ethics

Beyond moral and ethical considerations, making sure that AI is trustworthy and ethical can have an impact on a business’ bottom line.

A survey of technology leaders released by DataRobot in 2022 found one in three businesses (36%) experienced challenges or direct business impact due to AI bias, including lost revenue (62%), lost customers (61%), lost employees (43%), and even legal fees due to legal action (35%).

With regulators across the world viewing AI developments closely, it won’t be long before new regulations are announced. There are already a number of examples of governments setting laws to prevent AI bias and target rule-breakers with fines, including New York City banning the use of automated employment decision tools to screen job candidates unless the software has undergone a bias audit.

The Imperative of Ethical AI Practices

Companies are investing heavily in the promise of AI, with global business spending on AI set to reach $300 billion by 2026.

With proper AI usage, these advanced systems can not only improve business operations and profits but also to contribute positively to society, including opportunities to reduce environmental impact or improve accessibility for users with disabilities.

However, businesses should not simply adopt AI-backed tools without fully understanding their ethical implications.

Prioritising Transparency

Adherence to principles of fairness and transparency is critical when deploying AI tools. This should involve clear communication with users about how their data is being used, and even being transparent about how AI has influenced certain business decisions. Open-source AI models go some way in enhancing transparency, as these allow for third-party audits and a potential for greater accountability.

Ensuring Human Oversight and Fairness

Implementing human oversight for AI decisions can mitigate problematic outputs. Businesses may consider setting up cross-departmental committees that enable teams to raise their concerns. To enable this, businesses may need to provide AI literacy training.

Fairness should also be a key tenet of ethical AI use by businesses. This includes reviewing algorithms and updating those that could be discriminatory against certain user groups. External audits can also be put in place to ensure these fairness checks are unbiased and thorough.

Comprehensive safeguards must be established to detect and rectify any AI bias or unethical operations. This can be achieved through establishing ethical guidelines and accountability frameworks, using bias-detection tools, and ensuring that the data use to train AI systems is diverse and representative of as many backgrounds as possible.

Maintaining Data Privacy

Data privacy is a significant concern in the age of AI and digital technology. Businesses need to ensure robust data protection measures are in place and follow all relevant privacy laws and regulations.

If this data is not handled carefully, it could be accessed or misused by unauthorised parties, leading to privacy breaches and damage to the company’s reputation. Apple and Samsung have already made moves to ban applications like ChatGPT due to data leaks and privacy concerns.

Privacy concerns can also impact the adoption and effectiveness of AI applications. If customers do not trust that their data will be kept private and secure, they are less likely to use AI-driven services, which can limit the success and scalability of these applications.

Businesses should look to enhance their data privacy by anonymising data that they input into AI systems, minimising the amount of data retained on these systems, and ensuring data hygiene principles where possible.

As the world leans further into AI, businesses must prioritise the ethical use of this powerful technology. With careful planning and rigorous safeguards, AI can be a tool that is as ethical as it is revolutionary.

---