Businesses are struggling to implement "responsible AI" - but it could make all the difference

• Experian report finds 87% of leaders agree responsible AI will set them apart from competition
• Not even half feel they’re prepared to deploy AI responsibly
• Many lack high-quality data to take AI to the next level

Although AI tools have been proven to boost productivity in some cases, the technology is not without its fair share of concerns - namely job security, costs and emissions.

New research from Experian has found three in four (76%) businesses now agree that putting responsible AI into practice is now one of their biggest challenges.

This is despite 89% of UK business leaders acknowledging that AI is already improving their performance, and looking ahead, 87% agree that responsible AI will become a key competitive differentiator within the next two to three years.

How to deploy AI responsibly

Experian splits responsible AI into four core principles: reliability, privacy protection, minimizing bias and managing risks.

At the moment, companies are grappling with technical expertise (32%), applying AI to real-world use cases (31%) and balancing innovation speed with governance (30%).

Furthermore, only 45% have integrated responsible AI, with 10% lagging and 1% having no approach at all. Barely half (48%) believe their teams are adequately prepared to deploy responsible AI.

Experian UK&I’s AI and Automation General Manager, Christine Foster, summarized the key principles of AI: “Putting the right foundations in place, including high-quality data as well as clear accountability, and tools that support AI adoption across its lifecycle.”

Although nine in 10 agree that high-quality data is essential, only 43% feel confident in their data quality – this is Experian’s second step in its seven principles of responsible AI.

The report advises businesses regularly assess AI model performance; minimize potential risks to operations, people and customers; focus on security; put explainability tools in place; ensure privacy by design; and continually check for bias.

Some of the advice gems include starting small to prove value before scaling, running scenarios and tests in simulation before deploying, and diversifying teams involved in AI to broaden perspectives and reduce blind spots.

“As AI evolves, especially with autonomous systems on the rise, getting this right will be critical to building trust, enabling better business decisions, and staying competitive,” Foster concluded.

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!

And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.