Genative Artificial Intelligence (GenAI) is perhaps the most hyped technology of our time. Global investment in AI is expected to reach nearly $200 billion by 2025 as companies around the world prioritize AI investments to unlock productivity, efficiency and innovation.
Although everyone agrees that the technology is a game-changer for business, the question is increasingly being asked in the boardroom whether AI is really ‘enterprise-grade’.
Regional President EMEA at SAP.
Business AI is not new
While the hype surrounding AI has reached unprecedented levels, the technology itself is not entirely new.
Nick Bostrom provides an excellent summary of the advancement of AI technology in his book “Superintelligence”*, first published over a decade ago. This contemporary work captures the stages of AI: the inflated expectations; the plateau; and the breakthroughs.
Enterprises have long relied on machine learning to power advanced analytics and predictive capabilities across a wide range of use cases – from manufacturing to financial operations to purchasing and supply chain. These algorithms have provided business leaders with information to achieve greater operational efficiency.
AI has also been widely used in traditional forms of algorithms, for example in search engines, which have defined an entire era of our technological development and transformed entire sectors, especially the advertising industry.
Increasing issues with AI enterprise readiness
But what works on the internet doesn’t necessarily work in business. The internet doesn’t care about authorizations. The C-suite does.
As concerns about privacy and data protection grow – especially in light of continued regulatory pressure – many companies have implemented restrictions on the use of open GenAI tools.
This is not without reason. Imagine an employee sharing financial statements, vendor contracts, or payroll information with a GenAI tool that then reuses that information when answering questions from other users.
A GenAI tool without an authorization element is simply not at the enterprise level and is likely doomed to being limited to a single use case or department, limiting its ability to deliver value to the broader business.
Security concerns also arise with the concept of data lakes, which combine corporate data and external data sources for AI purposes. Data lakes can be treacherous for enterprises, especially when data needs to be exported outside large enterprise applications.
This requires a federated approach that leaves corporate data at the source and does not copy or transfer data. Critically, organizations must maintain the semantic layer of the data, which can be the Achilles heel of any data lake project and, consequently, of the GenAI models trained on that data.
Beware of hallucinations
The biggest danger of non-enterprise AI, however, lies in its tendency to hallucinate.
Gen AI is an excellent algorithm that learns fundamentally by looking at what is available within its domain, usually the Internet. Let’s be honest: you can no longer trust all the information on the internet.
In a corporate environment, CEOs are looking for the ‘single version of the truth’. This means that fact checking is important, but begs the question: “on which dataset should I train my Gen AI?”. The simple truth is that business leaders cannot build products or develop innovations using models that make things up or use insights based on false or inaccurate information.
This is where application suite vendors have the upper hand. The business applications that power the world’s enterprises have a wealth of business data that can be collected by AI algorithms to produce accurate, relevant, and reliable insights. Vendors in this space also have significant expertise in business processes and contextualized data – the perfect resources for training effective GenAI.
There is no doubt that companies will benefit from the power of AI in the coming years. Whether it is business-ready comes down to individual systems and tools. While some already have enterprise-level capabilities, others may not yet meet all the requirements for reliability and security. Business leaders must ensure they build AI use cases that can add value to the business, rely on robust data sets, and meet expectations. These guardrails will ensure business AI solutions that are relevant, reliable, and responsible.
We recommended the best AI website builder.
This article was produced as part of Ny BreakingPro’s Expert Insights channel, where we profile the best and brightest minds in today’s technology industry. The views expressed here are those of the author and are not necessarily those of Ny BreakingPro or Future plc. If you are interested in contributing, you can read more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro