[LI] Some thoughts on AI, intelligent agents, & transformation of the job market

Article originally published on LinkedIn, 28 May 2017

Source: “Long Promised Artificial Intelligence Is Looming—and It’s Going to Be Amazing,” Vivek Wadhwa, SingularityHub, Jun 17, 2016

So, if there were AI (artificial intelligence) bots or intelligent agents (IAs) to assist people looking for employment or a new job, how might they interact with increasingly intelligent processes used by organizations doing recruiting and hiring? And how could that change job seeking and hiring?

In a previous article on LinkedIn (“Towards AI for job seekers?“), I called attention to Robert Coombs‘ bot capable of finding a job listing and submitting a tailored application (which was in part a response to automation on the recruiting side). There are also other efforts to automate processes on behalf of job seekers, such as Esther Crawford‘s chatbot for handling recruiter follow-ups.

The potential to integrate various capabilities into an intelligent job search/application agent is exciting. However, in their current states of development, automation of the two sides of the job market – automating aspects of recruiting and hiring, which is an area getting more attention and investment, and nascent efforts such as those mentioned above to automate the job search – are not designed to talk to each other. Soon they will however, and then what?

The operant concept may be the IA – independent learning programs that will act on behalf of their owners, and learn along the way, although what operates for the job seeker and for the recruiter will naturally differ.

To the extent we begin talking about IAs interacting on their own – albeit with direction from and “ground truthing” with their owners – the dynamics become hard to predict. Keep in mind that these interactions would not necessarily be limited to job seekers with recruiters and vice-versa. IAs of job seekers could independently communicate with those of other job seekers (for instance to compare notes on particular employers), and recruiters with other recruiters (one start-up in Canada, Tabnex, is doing something in this latter area – again, another piece to fit into the larger picture), However there are some things that I think can be expected as IAs are rolled out and become more sophisticated:

  • IAs will be designed as learning programs, not just task-specific bots.
  • As IAs communicate directly with each other, this will not take take place in human language (if we take recent experiments as an indication of what is to come).
  • There will, therefore, have to be protocols and standards for communication among IAs, including ways to translate their communication to forms we humans recognize for checking and analysis.
  • IAs in this context, especially as used by individuals, would probably need a leash and a clock. That is some parameters for where to stop or not go, and for how long before receiving further guidance. Human input would be needed at major decision points.
  • Will IAs interact across the internet as is, or will dedicated virtual spaces be needed to facilitate? (Or in the post-internet neutrality world, will paid spaces be the only ones practical?)
  • In theory, IAs on both sides will be able to explore a much wider range of possible connections, especially in some sectors, than is possible with current applications of technology. For the job seeker, this would ideally include unlisted or not-yet-listed positions, anticipated by monitoring other parameters, such as investments and project grant awards.
  • The resume will no longer have a fixed form, except when needed for human reading, and then will change according to the context and the IA’s learning from experience – almost a quantum phenomenon. The database of one’s professional information and job history that goes into the resume will originate from the owner, and be tweaked as appropriate, but the specific selection and organization of information transmitted in each circumstance – or generated into a printed document in the appropriate human language – would likely differ.
  • Similarly, job announcements may change in form, perhaps in function of the range of potential applicants. And when an organization is hiring a team, it will be easier for the recruiters to assemble the right mix of talent and experience across the whole project activity, subdividing responsibilities dynamically according to those skill sets (rather than subdividing the tasks into defined positions before seeking people to fill them).
  • There will no longer be a need for companies to tell you that they’ll “keep your resume on file,” since recruiters will be able to page your IA (or the IA of anyone, or theoretically everyone) for resumes when they need them. Then again, recruiters’ intelligent programs may keep track of where to find interesting past applicants to speed that contact process, without precluding pursuit of any and all other contacts.
  • The design and programming of IAs for individuals will become a new market, possibly with a range of free, fremium, paid, and open source actors. IAs or other intelligent programs for recruiters and employers will likely be developed by the same class of companies already working in this area.
  • For better or worse, everyone can be on the market all the time, since IAs are designed to work independently.

Anyway, this is an attempt to see beyond the curve a bit about where automation and AI (artificial intelligence) might take the job market in the next few years.

(This post builds on some ideas written in “AI & the self-driving resume: 3. How AI might work for job seekers & recruiters,” 9 Apri 2017.)


Other blogs > LinkedIn > LinkedIn articles & posts, 2017 (Jan–Jun)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.