For CIOs, identifying and deploying artificial intelligence (AI) is one of the most discussed topics. In particular, how to balance the effective implementation of technology while understanding the issues related to its implementation, including bias, governance, integration and security.
There is now growing pressure on these CIOs and other technology leaders to act quickly regarding AI implementation. Organizations that are slow to act could be at a competitive disadvantage, and as a result, there is a struggle to balance fears of missing out and getting it wrong when it comes to AI integration.
Senior Vice President and Chief Information Officer at Juniper.
How accurate is AI?
Understanding the accuracy of AI is an important consideration for technology teams. For AI to produce accurate results, the analysis must be based on accurate and unbiased data. As new generative AI and large language models become increasingly popular, the industry as a whole has rushed to implement them. In turn, this has impacted the way AI is generally perceived, largely due to the negative focus on AI hallucinations and data biases used to train LLMs.
A recent study by Wakefield Research, a leading global provider of surveys, insights and research, found that 99 percent of participants felt that “some, most or all” AI data results were affected by bias. Additionally, 87 percent questioned whether it is possible to know if AI results are accurate.
In reality, biased results are inevitable if AI is fed previously discarded or obsolete data. Therefore, it is crucial to differentiate between various types of issues to assess whether and when bias may occur. To train LLMs there must be an evaluation of the uses of the data, how the AI is tested, and who uses the model. Techniques such as recall augmented generation (RAG) can be used to train AI and ensure accuracy.
Build trust in technology
To implement AI efficiently, CIOs and other decision makers must limit employees' blind trust in the technology. It is beneficial to see some acceptance and trust in new technologies from employees and team members. However, it is arguable that this degree of blind trust may be a more serious business concern than any possible hallucination or bias.
To limit the associated harms, it is essential that employees receive regular and up-to-date training on the different concerns and use cases of AI. According to Wakefield's findings, 84 percent of participants felt their employers should increase AI training. Business leaders should prioritize educating these employees, helping them make informed decisions when using AI in high- and low-risk circumstances.
An evolving AI landscape
Since AI and its capabilities have evolved so rapidly, it is understandable that IT professionals feel pressure to implement AI as a priority. In fact, 87 percent of respondents said they felt rushed to implement AI and 74 percent felt their company's policies couldn't keep up with the potential risks and benefits. Meanwhile, legal and business policies are often seen as obstacles to rapid implementation.
To increase return on investment, AI policies must be seen as a positive progression, rather than a regression. For example, some organizations have policies that address what data employees can and cannot share with third parties. Depending on the circumstances, companies may also rewrite policies to apply to external generative AI solutions. However, teams should be aware of the software purchasing policies and addendums for additional reviews of all AI-integrated solutions.
Workforce disruption
Like other innovative technologies, AI will inevitably change how the workforce functions today. However, this change will undoubtedly introduce new roles and skills into the workforce. This is reflected in 85 percent of respondents believing that AI will have some impact on creating future advancement opportunities.
For those looking to advance in an AI-enabled job market, individuals should prioritize skills related to tasks that cannot be performed using statistics, as AI is largely generated from statistics. The ability to improve both personal and collective effectiveness through AI tools will be highly valued. Workers who stay committed and continually hone their skills will find opportunities abundant.
Moving forward with AI
Even though AI has been around as a concept for several decades, the technology hasn't really entered the mainstream until recent years. Especially in recent months, AI has gone from something of science fiction to a fundamental business necessity.
Throughout 2024 and beyond, organizations without the right technology will risk being left behind. However, implementation must be done in a measured and sensible manner. Ultimately, stakeholders must fully understand how to use AI effectively to improve business operations, processes and employee well-being, without jeopardizing customer relationships.
We have introduced the best AI phone.
This article was produced as part of TechRadarPro's Expert Insights channel, where we feature the best and brightest minds in today's tech industry. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing, find out more here:
