CBA is testing the federal government's AI ethics principles
Commonwealth Bank of Australia (CBA) has revealed that testing the federal government's artificial intelligence (AI) ethics principles during the creation and design of its Commbank app feature Bill Sense, giving insight into how the bank could apply responsible AI at scale.
“CBA regularly uses AI to deliver features and experiences that are highly personalised, easy-to-use and add value for customers during every interaction. It was incredibly valuable collaborating with the government and other businesses on testing Australia’s national AI ethics principles in various situations,” said CBA Chief Decision Scientist, Dan Jermyn.
CBA is one of six organisations that has been working with the government since 2019 to better drive the adoption of ethical and responsible AI in Australia. They volunteered to trial a series of eight AI principles that were developed as part of the national AI ethics framework in real-world scenarios.
The eight ethics principles that were developed for the framework included: Human, social and environmental wellbeing; human-centre values in respect to human rights, diversity, and the autonomy of individuals; fairness; privacy protection and security of data; reliability and safety in accordance to the intended purpose of the AI systems; transparency and explainability; contestability; and accountability.
The ethics principles were developed following the release of a discussion paper earlier this year by Data61, the digital innovation arm of the Commonwealth Scientific and Industrial Research Organisation (CSIRO).
“Given the importance we place on using AI technology and the investment we’ve made into AI systems, participating and contributing to the government pilot was an easy decision for us. We see the government’s AI ethics framework as being essential when developing complex AI systems to help ensure AI is being used responsibly to improve financial wellbeing of customers and communities,” said Jermyn.
The Bill Sense feature in the CommBank was designed in accordance with the national AI ethics framework to help customers plan and stay in control of their regular payments and subscriptions up to 12 months in advance.
“Our well-established processes and governance, including around data safety and security, help CBA to apply AI safely and with the right accountabilities,” added Jermyn.
Beyond Limits: Cognitive AI in APAC
Courtesy of current estimates, it looks like Asia-Pacific AI will be worth US$136bn by 2025. Its governments and corporations invest more money than the rest of the world in AI tech, the data of its citizens is considered fair game, and its pilots are small-scale and, as a result, ruthlessly effective. This is why, according to Jeff Olson, Cognizant’s Associate Vice President for Projects, AI and Analytics, Digital Business and Technology, the APAC region ‘is right on the edge of an AI explosion’.
Now, startup Beyond Limits is pushing the boundaries of what AI can do, mirroring humans in its ability to find solutions with even limited information. As of this July, it’s partnered up with Mitsui, a global trading and investment company, to expand its impact in APAC.
How Does Beyond Limits Work?
Most AI companies claim that they can help businesses make better decisions. But many need astoundingly large stores of data to feed their information-hungry algorithms. Beyond Limits, in contrast, takes a different tack. Perfect data, after all, is largely a pipe dream kept alive by PhD students. In reality, systems must often make decisions from small, incomplete sets of intel.
But Beyond Limits’ AI is no black box. ‘When little to no data is available, Beyond Limits symbolic technologies rely on deductive, inductive, and abductive reasoning capabilities’, explained Clare Walker, Industry Analyst at Frost & Sullivan. While making these leaps in logic, however, the system also keeps track, ensuring that humans can review the AI’s ‘thought process’.
Why Partner With Mitsui?
Beyond Limits is built for specific applications such as energy, utilities, and healthcare—but lacks the extensive industry network of Mitsui. Partnering allows Beyond Limits to access a portfolio of firms specialising in minerals and metals, energy, infrastructure, and chemicals. ‘We’ve been working on this deal for several years’, said Mitsui’s Deputy General Manager Hiroki Tanabe. ‘Mitsui’s global portfolio and Beyond Limits’ AI technology will...deliver impact’.
In the first test of that dramatic statement, Liquified Natural Gas (LNG) will soon deploy Beyond Limits’ new system. If everything goes according to plan, LNG will optimise how it extracts and refines energy, making money for both itself and investors—including Mitsui. This, in fact, is Mitsui’s strategy: go digital and don’t look back.
Why Does This Matter?
Forty-five percent of Asia-Pacific companies surveyed in Cognizant’s thought leadership ebook consider themselves AI leaders. Positivity bias, that oh-so-common tendency of humans to position themselves as above average as compared to others, strikes again. (Most small companies fail to launch successful AI projects on their own.) And partly, this is because firms fail to integrate AI with industry expertise.
‘A large part of the focus on talent for AI today has been getting the people who are strong in mathematics, AI, and technologies’, said Olson. ‘But where you make your money out of AI projects is when you apply them to your business’. In short: APAC nations looking for ways to bridge the gap might follow Beyond Limits and Mitsui’s playbook—coupling startup AI with a corporate network.