In this week’s newsletter, Lee Bryant shares some links relating to the emerging debate about the transparency and role of algorithmic sharing of information in the digital workplace.
In the wake of last week’s Facebook / CA furore, people are starting to question the role that algorithms play in shaping our thinking and behaviour on social channels. Facebook’s switch from a purely chronological newsfeed to one that decides what you might ‘like’ to see frustrates some users (and they are not the only platform doing this), but I don’t think we really yet know how it shapes our thinking or contributes to keeping us locked in our comfortable filter bubbles.
Information overload is a growing issue on public social platforms like Facebook and Twitter, but also in the workplace, where the rise of tools like Slack and Microsoft Teams are increasing both the speed and volume of information sharing. We can’t read it all, so do we let it flow by and hope network amplification will help us filter the river, or should we rely on algorithms to identify what we need to see, which is not the same as what we want to see – yes, I might ‘like’ another funny cat video (or boring CEO update), but perhaps what I need to see is the really obscure new engineering spec being shared by someone who rarely engages with me.
If we go down the road of algorithmic sharing in the enterprise, as I think we will, then how can companies and individuals have some oversight or control of the algorithms that are used, and what kinds of outcome should they be optimised for?
In the wake of last week’s Facebook / CA furore, people are starting to question the role that algorithms play in shaping our thinking and behaviour on social channels. Facebook’s switch from a purely chronological newsfeed to one that decides what you might ‘like’ to see frustrates some users (and they are not the only platform doing this), but I don’t think we really yet know how it shapes our thinking or contributes to keeping us locked in our comfortable filter bubbles.
Information overload is a growing issue on public social platforms like Facebook and Twitter, but also in the workplace, where the rise of tools like Slack and Microsoft Teams are increasing both the speed and volume of information sharing. We can’t read it all, so do we let it flow by and hope network amplification will help us filter the river, or should we rely on algorithms to identify what we need to see, which is not the same as what we want to see – yes, I might ‘like’ another funny cat video (or boring CEO update), but perhaps what I need to see is the really obscure new engineering spec being shared by someone who rarely engages with me.
If we go down the road of algorithmic sharing in the enterprise, as I think we will, then how can companies and individuals have some oversight or control of the algorithms that are used, and what kinds of outcome should they be optimised for?
For some further reading on how algorithms are shaping our environment, we’ve put together the following links:
- Few laws govern how data are collected at work, in the age of AI where does this lead? From The Economist.
- Michael Natusch explains financial services firm Prudential’s strategy around AI – focusing on causation and correlation.
- An interesting article outlining the risks and mitigations of content production algorithms inside organisations with a focus on Microsoft Graph.
- “AIs choose answers. Humans choose questions.” Given all of the possibilities of technology, the next question for us to choose is: What’s next?
- How organisational network analysis enabled a transition to a more agile enterprise.