There are seismic changes coming with the rise of generative AIs such as ChatGPT from OpenAI.
You only have to look at the response of global governments for evidence of this. UK Foreign Secretary James Cleverly recently called for coordinated global action to ensure responsible development of artificial intelligence while chairing a United Nations Security Council session, while his government plans to bring world leaders together this autumn for the first major global summit on AI safety.
Kit Cox, founder and CTO of workflow orchestration platform Enate, says the pace of change “is the most rapid technological advancement I’ve seen in my lifetime”.
As businesses increasingly turn to automation to perform or assist with menial tasks, as well as to augment more complex projects, so have fears grown over job displacement.
“The possibilities of AI are literally changing week by week,” Cox tells BusinessCloud. “When it comes to the economy, it’s evident that certain jobs will become obsolete and certain roles will undergo disruption.
“It would be irresponsible for a founder of a technology company to claim that there won’t be any impact on employment… it is crucial that we provide support to individuals navigating this change.
“Throughout history, technology has consistently eliminated and created jobs. However, the positive aspect is that currently there are more job opportunities available than the workforce can handle.
“Machines can only replace humans up to a certain extent; human involvement remains essential. Nevertheless, it is evident that our society requires increased productivity in the market to enhance the functioning of the economy, and artificial intelligence can assist in fulfilling this need.
“Rather than replacing jobs entirely, many positions will become AI-enabled.”
He adds: “While the concept of Universal Basic Income is being discussed, I have concerns regarding the potential risks it poses, such as regressing to a state of control reminiscent of the Middle Ages.”
Despite the grand words and gestures, Cox suggests that the government does not fully appreciate what is coming. “There is a real naivety in the UK government about what AI is capable of,” he says.
“Although it is generally agreed that legislation needs to be in place, there seems to also be the opinion that ‘they’re just large language models, so they’re not capable of doing things, therefore, don’t worry’. “Storytelling, and the power of misinformation is where the greatest existential threat comes from. If we do not successfully regulate through for example, watermarking, we will have a significant problem, most likely the first instances of this will be the democratic elections which are coming up in a number of Western societies… misinformation and deep fakes are the most pressing issues.”
London-based Enate, which today employs just short of 100 staff, was founded in 2013 to give business leaders full visibility over their operations, and let them manage all work in one place.
“Too often, processes are scattered in various pockets, making it hard to see the bigger picture and identify problems,” explains Cox. “We solve this with one end-to-end platform to manage everything.”
Users can design and configure business processes through its interface – creating workflows and defining roles and responsibilities for both the human and the digital workforce. It integrates with a variety of robotic process automation tools, enabling collaboration between human workers and software bots, and monitors performance.
Customers include enterprise businesses EY, Infosys, CapGemini and TMF Group. The latter, a multinational professional services firm operating in over 50 countries, recorded a £32 million margin improvement and 22% improvement in efficiency by using Enate’s operations management platform, according to Cox.
Last year the business secured $2m of funding with Mercia. Cox says it is “now open to additional funding rounds to fuel our next phase of growth”.
Enate recently launched EnateAI, a plug-in powered by GPT-4 – a multimodal model which ChatGPT maker OpenAI says “exhibits human-level performance on various professional and academic benchmarks”. GPT-4 accepts image inputs as well as text prompts.
EnateAI can automatically perform sentiment analysis, categorise emails, understand foreign languages, automate queries and extract email data/auto-populate forms, according to Cox.
“AI has completely changed the game of what’s possible. For example, before GPT-4 if a business wanted to automate email triage they would have to incur a high financial, time-intensive and technical cost to implement this,” he explains.
“We are talking hundreds of thousands of pounds, with typically only 70% accuracy. In the example of automated email triage, you’d have to collect at least 30 examples of each type of customer email a business wanted to classify and then iteratively train a model. This model then had to be maintained and if a new classification was later added, the whole implementation had to undergo rigorous testing.
“Now, with GPT-4 technology, AI-enabled email triage can be integrated seamlessly and platforms like EnateAI can automatically scan the contents of any email to determine the most appropriate defined category.
“This removes the need for complex MLOps, high costs and long training lead times, and is around 90% accurate compared to the pre-GPT-4 days of 70%.”
Inevitable CEO Ben Grubert will discuss the rise of AI, the evolving regulatory landscape and offer practical tips for harnessing ethical AI into your business in BusinessCloud’s Demystifying Tech podcast webinar. You can sign up at the link above
On the general future direction for AI, does Cox believe we will see a few massive platforms dominate the space, or a wider pool of specialist AIs?
“In my view, certain older technologies will undergo a swift transformation from being costly and valuable to becoming readily available commodities,” he says. “Take, for instance, intelligent document processing: currently an expensive venture, but it’s likely to transition into a commodity soon.
“The integration of general AI models into practical tasks, beyond pure information usage, holds immense potential for groundbreaking advancements. The game will truly change when these potent AI models can seamlessly access other systems, transitioning from mere reasoners to active generators, unlocking new possibilities and opportunities. For example, the ability to book a holiday, purchase car insurance or create an end-to-end website using AI.”
Asked for tips for businesses looking to harness AI effectively, he answers: “To effectively train and utilize AI models, data sharing is essential. Therefore, you need to establish well-defined data management rules that outline what can and cannot be shared with both public and private AI models.
“Ensure your entire team is familiar with and adheres to these guidelines. Depending on the size of your organisation, it’s sensible to think about hiring a chief AI officer to take responsibility for AI safety.
“Initiate a collaborative brainstorming process with your staff and seek input from an external expert to explore the vast possibilities of AI. Keep in mind that the landscape of AI capabilities is rapidly evolving, with new possibilities emerging weekly. This era marks the swiftest pace of technology-driven change in memory.
“Categorize your identified use cases into four primary areas: skill support, transaction support, customer support, and decision support. It is crucial to recognize that all these use cases are designed to assist and empower humans in their tasks and find the correct AI co-pilot to help with these areas. By leveraging AI effectively, you can enhance human performance and productivity.”