Last week, at the inaugural conference of AI on a Social Mission, organized by Alliance Impact Intelligence Artificielle (AIIA), in Montreal, Canada, I shared my views on the policies and legal implications of artificial intelligence on the world of work.
Panel discussion on “Our Laws and Our AI” at the AI on a Social Mission | AIIA
One of the great uncertainties surrounding artificial intelligence is how it will impact the world of work.
For example, this is a masterpiece by Picasso, the Large Bather.
There’s just one problem when trying to share this masterpiece online, the Facebook and Youtube filtering algorithms cannot tell whether it’s art or porn.
This is why social media companies hire Commercial Content Moderators (CCMs) to act as “digital trash removers” when their algorithms fail to do the job.
Approximately 100,000 people work as content moderators. They often work more than 10 hours a day, in poor working conditions, sifting through graphic images, videos, and texts. As one might imagine, their job can have adverse psychological effects, such as depression and paranoia.
With the help of AI, new forms of work are being created, but not all are desirable.
This is just one example of how AI is transforming the world of work with immense regulatory and societal implications.
I co-manage the Technology@Work Initiative at the International Training Centre of the ILO, the training arm of the International Labour Organization.
The Technology@Work Initiative is a supporting platform for the Future of Work Centenary Initiative. The Future of Work Centenary Initiative is the centerpiece of the ILO’s centenary activities to mark its 100th anniversary in 2019. I welcome to engage us on our Technology@Work platform and check out the Centenary Initiative.
We need AI to help us solve global challenges that are too complex for us to solve alone, including the Future of Work and the achievement of the Sustainable Development Goals (SDGs), like SDG 8 of decent work for all and inclusive economic growth.
AI’s ability to recognize patterns, optimize processes, and provide insights is facilitating disruptions in the organization of work and production around us.
AI is enabling disruptions with.:
All these emerging changes in our economy will require better foresight, so that we can be proactive in shaping the policies and laws to mitigate the risks.
How have work and production been disrupted?
Today, AI is enabling the emergence of new forms of work, including digital labour, such as the work Content Moderators perform. These new forms of work are often more flexible but also more insecure and precarious, widening existing inequalities.
The rise of the gig economy as a new business model often powered by algorithms to optimize pricing and allocation of resources on Facebook and Amazon Mechanical Turk is also promoting the casualization of workers. This is making workers more disposable, shifting the fundamental rights of workers and the responsibilities for the employees away from the employer like Uber.
There will be fewer winners and a lot more losers. AI is also enabling the rise of the winner-takes-all economy, increasing inequality and income disparity. Middle-level jobs, like in manufacturing, disappear diminishing the middle class.
A key part of this winner-takes-all economy is the concentration of digital power in a handful of AI giants or super corporation like Google and Facebook. Due to their first-mover advantage and size, they can distort markets and buy out competitors. In 2015 Google captured over half (55% or USD$ 44.5 billion) of global digital advertising revenue. Its next closest competitor was China’s Baidu, at 8%.
Some may ask if we need better enforcement of existing laws and update current antitrust policies and regulations to prevent the winner-takes-all economy.
Moving forward, we will need more broad-based public dialogues, like this very conference, to ensure the fundamental rights of workers and influence new market regulations to guard against exploitative new forms of work and the formation of the winner-takes-all economy that is facilitated by AI.
How can AI help us build the Future of Work we want? Will AI take away jobs by substituting human labour for capital?
Based on historical evidence the fears of mass unemployment due to technological change has been proven to be unfounded. Rather than eliminating occupations, technology will change the way we complete tasks. The number of jobs replaced by AI will be offset by new job ones in the medium to long term.
By designing and applying AI-enabled business models and processes which compliment workers we can ensure that the dividends of AI and innovation are shared more broadly.
Another promise of AI is in the tracking and forecasting of labour market information to anticipate the competencies and skills needed in the future. The rapid rate of automation enabled by AI will make newly acquired skills and competencies become obsolete faster. This calls for more adaptability from individuals and institutions, and the promotion of lifelong learning.
To end, the Future of Work we want belongs to all of us.
Technology does not determine our future. Together, we must imagine and give shape to an ambitious Future of Work we want.