From AI in the military to intelligent traffic systems, it’s another bumper week in the world of AI. Here’s what you need to know today:
Cisco Take On Alexa With New Business-Friendly Voice Assistant
Siri, Cortana, and Alexa could soon have to move over in the office, following the release of Cisco’s AI assistant for business.
Yesterday saw the first live demo of Cisco’s Spark virtual assistant, first announced back in November 2017, at a Cisco Live event in Melbourne, Australia.
Cisco senior VP and general manager of applications Rowan Trollope told the event that “existing voice assistants that have traction – none of them were built for business. They were built for the consumer domain. That is a totally different domain space to the one we are interested in.”
The company believes it can combine its strong positions in collaboration hardware and software into an enterprise AI product.
“We think there is a great opportunity to do that kind of AI in business,” Trollope said. He went on to mention that Cisco’s partnership with Apple could see Siri and Spark eventually collaborate with one another.
Google’s AI Is Being Used By US Military Drone Programme
Google’s TensorFlow AI framework is being deployed by the US military, in a highly controversial world-first for AI in combat.
The Department of Defense’s Project Maven, established in July of last year, uses machine learning and AI to analyse visual data shot by US drones overseas. The neural network is able to analyse a video feed, detect objects of interest, and flag them for a human analyst to review.
Drew Cukor, chief of the DoD’s Algorithmic Warfare Cross-Function Team, said in July, “People and computers will work symbiotically to increase the ability of weapon systems to detect objects. Eventually we hope that one analyst will be able to do twice as much work, potentially three times as much, as they’re doing now. That’s our goal.”
It might not be killer robots, but the revelations about Google’s collaboration with the military have nevertheless proved controversial among the firm’s staff. While the corporation has worked with government agencies for years, the move to directly aid Project Maven has caused much internal debate – with some staff reportedly outraged at the use of the company’s AI.
“Military use of machine learning naturally raises valid concerns. We’re actively discussing this important topic internally and with others as we continue to develop policies and safeguards around the development and use of our machine learning technologies,” a spokesperson for Google said.
Modern AI Means A ‘Collaborative Universe’ Of Bots
AI might be nothing new, but the rise of a ‘synthetic collaborative intelligence’, in which modern AI systems are able to communicate with one another automatically, is revolutionary. That’s the opinion of one Mike Duke, Chief Innovation Architect for Wells Fargo.
“I’ve been involved with AI since my college days back in the 80s but never have I seen it begin to collaborate with itself,” Duke told AI Business. “It’s finally landed in living rooms, and now it’s going to become part of people’s daily lives. We’re going to be participating in this new universe of different bots.
This is demonstrated by a proof-of-concept HR tool called Socrates that’s currently being trialed by Wells Fargo. “We envision that, someday, team members will be able to say to their home Alexa, ‘I’d like to take next Thursday off’. Alexa will communicate with the internal HR chatbot at Wells Fargo, get the boss’ permission, and then come back to you and approve the request.”
The potential for this collaborative form of technology are manifold for businesses, as outlined in Mike’s interview: “Inevitably, these bots are going to have a collaborative network that better understands our customers and team members so it can provide a better and more secure experience.”
Have We Reached Peak AI Hype?
Some warn that the end is in sight for AI hype. According to McKinsey, there were twice as many articles referencing AI in 2016 as in 2015 and four times more than in 2014 – a trend that has continued into 2018. This conflicts with Gartner’s hype cycle, which estimates AI hype has another two to five years before it reaches its peak.
“What will ensure AI’s success and help it emerge unscathed from the hype cycle will be plugging AI into the very fabric of every enterprise and making it part of the decisionmaking process. This will be a particular challenge because it will require companies to make a fundamental change to how they think about business and how they build strategies,” Roman Stanek, Founder and CEO of GoodData, argues in Forbes. “However, it will also require companies to think about how they will overcome employee and customer skepticism before they can reach the ultimate end goal of automating processes, whatever those may be.”
The end of AI ‘hype’ might not be such a bad thing, and could indeed be reflective of the increasingly pragmatic approach companies and stakeholders are taking to the technology as its use cases become apparent in everyday life. Anything that ends the scourge of The Creation of Adam pastiches and doomsday headlines is surely a good thing.
Machine Learning Moves Traffic In Smart Cities
Smart cities are about to enter a gold-rush era, thanks to pioneering intelligent traffic systems that quickly learn about traffic patterns using machine learning.
As old infrastructure crumbles and regeneration projects abound, cities are already taking advantage of AI technologies to transform themselves. Cities like San Jose and Dallas are already experimenting with intelligent traffic systems in public-private partnerships, offering useful case studies for AI in transportation – and a glimpse of our fully-automated, self-driving future.
“The idea of using existing transportation networks as efficiently as possible makes total sense,” says Vijay Sammeta, CEO of CivicFoundry. “If you’re a city mayor thinking about a new traffic control system, I can’t imagine you’re not thinking of it connected in some way.”
Machine learning tools collect traffic data from many sources, including radar images, historical surveys, and Internet of Things (IoT) sensors embedded in roads and traffic lights. The intelligent traffic system is then able to dynamically adjust the signal timing of traffic lights based on this data.
“This is where AI and machine learning really shine,” Sammeta argued. “They’re able to do these calculations much faster, run through all the permutations, and play what-if scenarios in a manner that is so much more efficient. This is the perfect use-case.”