As corporations plan out their generative AI methods with duty in thoughts, one of many sticking factors is learn how to begin. To assist them, corporations have begun releasing accountable AI platforms to information decision-makers in approaching the know-how. 

Accenture and AWS are the newest corporations to supply a method for corporations to check their AI readiness and consider AI purposes. The brand new providing is known as Accenture Accountable AI Platform powered by AWS.

Accenture Chief AI Officer Arnab Chakraborty advised VentureBeat the platform offers a great begin for corporations simply now interested by accountable AI whereas providing flexibility for anybody additional alongside the journey. 

“Accenture and AWS are bringing collectively the perfect of each corporations and the capabilities we provide, not simply to corporations in isolation however with our ecosystems,” Chakraborty mentioned. “Now purchasers have entry to the totally different device stacks that natively come from AWS and companions, which supplies flexibility to the shopper.”

Accenture labored with AWS to supply the platform to their mutual purchasers. AWS Vice President for Generative AI Vasi Philomin mentioned Accenture’s accountable AI platform works complementary to the accountable AI and AI security instruments provided on AWS. 

In a demo to VentureBeat, the platform guides customers from doing a listing of the corporate’s AI purposes to evaluate its AI maturity and submit purposes for compliance. It would primarily work for corporations utilizing AWS to run AI providers. Customers also can customise what the platform checks for. If the corporate needs to emphasise one space of accountable AI or focus particularly on dangers for his or her business, they’ll accomplish that. 

Chakraborty mentioned the concept for the platform is to information corporations all through the complete lifecycle of AI applications. The platform helps corporations determine particular dangers, decide learn how to tackle these points and create compliance applications for any altering rules. This course of helps enterprises to scale up or down any generative AI pilot applications. 

AWS already presents accountable AI and security applications, together with spinning out AI security platforms as its personal purposes. One such utility is Guardrails, which AWS lately launched as a standalone API. AWS additionally partnered with Anthropic to assist scale gen AI extra responsibly. 

Accountable AI efforts are rising

Surveys have proven that enterprises consider accountable deployment of generative AI is essential. Nevertheless, a current one from PwC famous that solely 58% of 1,001 corporations have even begun addressing it. Accenture’s personal analysis discovered that solely 2% of corporations are answerable for AI operations of their companies. Accountable AI primarily appears at dangers round hallucinations, bias and typically security.  

One such purpose, Chakraborty mentioned, is that corporations are sometimes already overwhelmed with constructing gen AI purposes, so including accountable AI duties may be advanced. It’s not that enterprises don’t consider it’s important; they do. In addition they perceive the real-life issues that may start ought to the AI go awry. 

Nevertheless, different hindrances for enterprises nonetheless exist. This may be wherever from agreeing on what accountable AI even means for the corporate, to confusion over who leads the cost, lack of expertise and prioritization. Whereas suppliers like Accenture and AWS supply technical platforms for enterprises to start out a path towards accountable AI, there nonetheless must be extra steering for companies to know what accountable AI appears like absolutely and means for them. Enterprises should transfer quick in order that the know-how doesn’t evolve sooner than accountable AI insurance policies.