“Sinequa is differentiating itself through its use of deep learning (artificial neural networks), by applying multiple deep learning models to provide more accurate and relevant search results,” writes Alan Pelz-Sharpe of Deep Analysis in a recent report.
It seems only yesterday that cutting-edge technology in the legal sector was defined by souped-up search engines (e-discovery) and document management systems (matter management). Genuine innovation and true high technologies seemed out of place in law firms. Even today, many law firms remain paper-centric and look askance at things such as cloud computing and AI. But a changing of the guard is underway as law school graduates from the late 2000s now become partners and general counsels. In short, Millennials are taking over, and with that change comes a sudden and jarring shift in attitude toward technology. Far from rejecting it, younger lawyers are embracing it with open arms. Witness the recent Legal Week event in New York City. Though always a big conference, it typically has been dominated by dated and dowdy technology vendors; this year, it was awash with sellers touting the wonders of AI.
For those choosing inaction, on the other hand, Alan Pelz-Sharpe, founder of analyst firm Deep Analysis warns, “In a world of simple to access and use tools to doctor images, it’s all too easy to defraud. The risk and regularity of this kind of fraud is likely low today, but it will certainly increase substantially over the coming years.” So while all may be quiet for the moment, the cost of inaction may be high. “In all likelihood, few if any insurance firms have addressed this growing concern. Yet it should be a priority for them as once this takes off — and it will, and it will be hard to stop,” said Pelz-Sharpe.
These small startups are saying that it doesn’t make sense to always use that sort of sledgehammer approach of deep learning. By taking a very narrow approach, they’re able to train on a much smaller data set, be up and running much more quickly and, frankly, be much more accurate. “—Alan Pelz-Sharpe, founder, Deep Analysis
Industry analyst Alan Pelz-Sharpe shared his thoughts on this topic in a recent KMWorld article. For starters, Pelz-Sharpe summarizes the all-too-typical corporate information environment this way: “We have multiple disconnected silos of information, mountains of duplicate files, poor search functionality, and often little-to-no structure or governance applied to knowledge or information assets.” Sadly, this is the case at most organizations when they first approach Northern Light to explore our enterprise knowledge management solution for market and competitive intelligence. It is not uncommon for such organizations to have a glut of uncoordinated Microsoft SharePoint sites on their network – one pharmaceutical company that came to us had 50,000 SharePoint sites… and another 80,000 abandoned SharePoint sites! – which is a recipe for information chaos.
Today, we live in a world of cognitive content services, ambient search and hyper-automation, with multiple sects, gangs and subgroups lurking around the corner from RPA to insight engines. All this effort to label, often entirely different products, into a single category helps nobody or very few. Take, for example, all the enterprise search tools in the market today. It’s in our nature to simplify things, but the ambiguities and differences between, say, the Microsoft Azure Cognitive Search product, ElasticSearch, Searchblox or Sinequa are essential to understand. They are all enterprise search products, but they are all quite different and incredibly complex.
“This constant adding of new functionality and not raising the price of the subscription is unusual in the software world,” Pelz-Sharpe said. “It is common with Microsoft, common with Google, common with Oracle. Box is a very healthy company, but it’s not Oracle, right? Many vendors, when they get to [Box’s size], end up with a price list.” The security features of Box, Pelz-Sharpe added, have made the company popular among government users as well as law firms.
Last year my firm Deep Analysis ran a survey that asked 500 people working in U.S. software firms about their work from home and hybrid work experiences, focusing on collaboration systems. Back in the day, meetings had agendas, minutes were taken, and action items were assigned to individuals. Pretty basic stuff, if a pain in the backside to do. Even so, most meetings were at least focused, action-oriented and generally productive. Today, many online meetings are none of the above, with multiple attendees wondering why they are there at all. Long story short, a majority of people in our survey were fed up with dealing with the numerous pointless and confusing meetings.
In other areas, the White House is adding $187 million to the National Institute of Standards and Technology for research initiatives and the development of standards to accelerate the adoption of emerging technologies such as quantum science, advanced biotechnologies, and artificial intelligence. But the funding increase is too little, too late, said Alan Pelz-Sharpe, founder of market research firm Deep Analysis. Pelz-Sharpe said it would be good for the U.S. to set standards for AI, particularly around the areas of fair use, transparency, and bias. However, he said the U.S. is behind the curve when it comes to standards-setting and likely won’t set a precedent for the rest of the world to follow. The international community has already moved ahead to define and set standards. Alan Pelz-SharpeFounder, Deep Analysis “The international community has already moved ahead to define and set standards,” Pelz-Sharpe said.
Learning from China’s approach Though it may be “counterintuitive” to give China credit regarding its AI regulations given the political climate, Pelz-Sharpe said its rules “set a new benchmark that may drive improved regulations elsewhere.” The AI regulations outlaw algorithmic price gouging and require an explainable algorithmic decision-making process. The algorithms have to be “trustworthy AI,” meaning built and tested to be as fair, explainable and bias-free as possible, Pelz-Sharpe said. While there are discussions about trustworthy AI in areas such as Europe and the U.S., Pelz-Sharpe said there’s little from a regulatory standpoint when it comes to defining the phrase. China, he said, is “making this concrete.” “Beyond blatant and deliberate misuse of AI, two of the biggest factors of concern for its use are bias and explainability,” he said. “These regulations at least attempt to address some of these concerns.”