Artificial intelligence and machine learning technologies are making the world around us more efficient and easier to navigate, freeing up our time as humans to focus on the things we do best.
However, when it comes to the business world, these technologies often focus more on accuracy, algorithms and statistics—not what businesses actually need to succeed. Human beings and real-life problems are more than algorithms punched into a data system, and technology sometimes struggles to realise this.
“When it comes to the current uses of AI, all of us obsessively talk about accuracy, but accuracy is not the point” Aible CEO and founder Arijit Sengupta told Forbes. “The point is impact.”
To make the world of AI more accessible—while incorporating real-world solutions, factors and obstacles into the equation—AI startup Aible has developed Aible Advanced, the first-ever solution to fully automate machine learning.
The platform also has a version for business users, helping companies automate the machine learning process to get real results for unique business situations. No data science training is needed—just a business problem to solve and enough curiosity to solve it.
The concept behind the AutoML programme is to allow both data scientists and business users to use their strengths to find the best predictive model for their needs.
While the version of Aible Advanced for developers and data scientists automates the most repetitive parts of data science to help users focus on their work, the version for business users removes the need to understand data science at all.
“Business users know their business,” Arijit stated. “Let’s teach the AI to speak business, rather than teach business users to speak AI.”
The platform understands the goals and obstacles in business, then creates a customised prediction model to maximise business impact.
“Business users have the business domain knowledge; data scientists have the modeling expertise,” Arijit explained in a statement. “Aible Advanced, in conjunction with Aible, lets each type of user seamlessly contribute their unique skills and knowledge to generate the best predictive model for their unique business reality.”
As someone highly proficient in AI, Arijit eventually realised that these programmes have a tendency to be “accurately wrong.” He also stated that the human experience is often minimised for the sake of technology.
“I really felt people almost disrespect the human too much when we evaluate AI,” he told Forbes. “We get so excited by the technology, we forget that human beings have value.
“If you ask a data scientist how should I decide whether I should deploy this model? They’ll say, ‘This model has 85 percent accuracy, it is so much better than random, (in fact, sometimes they will say it’s so much better than a coin toss) you should deploy it.’ The problem is when was the last time you saw a bunch of salespeople sitting there flipping coins trying to decide what to do?”
Forrester called Aible Advanced the “best choice for pure businesspeople” in The Forrester New Wave: Automation-Focused Machine Learning Solutions Q2 2019.
“Unique among AutoML users, Aible gets that a model that maximises accuracy almost never maximises business impact,” Forrester said.
Robert Scott Lazar
Join the discussion