Smart agents will reset the working relationship between people and technology.
The working relationship between people and technology is fundamentally shifting. Employees used to learn the language of business applications just in order to use them. A world where ‘Shift F9’ created a new line item entry in their expenses.
During the first PC revolution in the 1980s, the time of Apple II and the original IBM PC, forward thinking parents sent their children to computer camps where they learned to write programs in the BASIC language. They hoped to give them the chance of a bright future in the incoming era of computers.
Fast forward to today people still need to learn how to use digital devices, but advances in user experience design have masked the language of technology with buttons and guided navigation. The user training toll has decreased but it has by no means gone away. We now require employees to engage with more applications than before, from HR systems to self-service analytics. Each user interface is more sophisticated and easy to use but is also likely to have its own user experience and language.
Smart agents in the future
In the future, smart agents will remove the burden on organisations to train up ‘tech-literate people’ by delivering ‘people-literate tech’. Employees will rely on smart agents they control to interact with applications for them. They will converse with these smart advisors, or virtual personal assistants, in their natural language via touch, speech, keyboards, written text, gestures, and other more human mechanisms. In the future, an interaction with business applications might begin with a blank screen and a question mark.
In time, with their owner’s permission, these smart agents will both gain more insight into what their owners want and need through observing what they do and also proliferate across devices and interfaces delivering devices to the user anywhere and everywhere.
By 2020, we predict that at least 40 per cent of people will primarily interact with people-literate technologies like these, removing much of an organisation’s need to invest in computer literacy training. CIOs should be planning on people-literate technologies becoming the dominant model and have smart agent trials on their roadmap.
Already, we are seeing evidence of smart agents driving this transformation, not just with conversation aids like Siri and Cortana. Expert systems like 'Ask SGT Sta' – a chatbot developed by Next IT for the US Army – already has over 800 rules to determine what to say to users. IPSoft’s Amelia offers virtual customer assistants that learn from observing person-to-person interactions, and IBM’s Watson offers a broad suite of natural-language processing and generation capabilities.
Natural language and intelligent interaction is just the start. Smart agents will become proactive, autonomously making suggestions based on what they learn about the people that use them and what’s in their diary, on their to-do-list, in a business workflow or happening in their personal and professional networks. They will actively learn from responses and, with permission, begin taking actions.
As smart agents grow in popularity, demand for ‘bring-your-own’ smart agents will follow. Various factors however, may slow, or speed up, the arrival of these and new technologies in the workplace.
Early smart agents are still reliant on people-curated, rules-based expert systems to understand the meaning of what’s been said, draw inferences and decide on how to respond. These tools are cute, sometimes fun and sometimes useful, but they’re not yet smart enough.
Over time, CIOs should be developing an organisation-wide view on the potential for smart things. They must remain sensitive to the fact that people will not automatically trust smart agents and that individual business units may need persuading of their potential.
Trials will be key. A number of Gartner clients are already evaluating the value of people-literate technologies. This establishes the spirit of a digital workplace, while allowing employees to have greater input into how smart machines are adopted.
Tom Austin, VP and Gartner Fellow