“AI is the most important issue shaping society,” said veteran venture investor Ted Dintersmith.
Our new paper on the implications of artificial intelligence concluded, “Code that learns will prove to be humankind’s greatest invention.”
Concerned that the country wasn’t mobilizing for the automation economy, Ted left his fund, and produced a movie, Most Likely to Succeed, to name the problem. He visited all 50 states hosting conversations about what’s happening, what it means, and how to prepare.
“If we don’t address the problem, we’ll have a lot of angry, alienated people on our hands. It’s all related, the pattern of innovation, school, and democracy,” said Dintersmith.
At SXSW EDU, we hosted a meetup to hold a quick version of Ted’s conversation.
Richard Boyd, CEO of Tanjo.ai (on the right above), facilitated a conversation with some great technologists (including Google’s Jonathan Rochelle, on the left above). They discussed the 50 year history of AI and the remarkable rise in the last three years driven by more devices (mobile, sensors, cameras), powerful computing and cheap storage.
New machine learning tools, at a minimum, can read really fast, recognize patterns and apply rules— and they never forget. For example, Boyd’s machine learning engine read every U.S. legal opinion since 1797 and produced a complete correlated map in less than a week (below)
Machine learning systems can read everything written by a person from history and construct a good model of that person. Boyd has investigated the works of leading scientists and constructed a number of “animated persona” to promote STEM learning. Soon learners will be able to engage with historical figures in natural language conversations.
For the next decade nearly every job will become more augmented with smart tools. The technologists discussed shaping the right balance between humans and automation to optimize desired outcomes.
Sarah Klein, Cvent, explained that we all leave a big data exhaust that, in many cases, is being collected. She asked, “What are the ethics for collecting data if we don’t know what we’re going to do with it yet.”
On Blockchain, the technologist said that beyond the cryptocurrency hype, there is real hope for better portable transcripts and privately managed learner profiles.
Boyd pointed to the potential for better privacy through blockchain, by eliminating the need for big databases like the frequently hacked credit rating agencies.
What does it mean?
Russ Altenburg (@RussAltenburg), CEO of Reframe Labs (@ReframeLabs), led a discussion of the civic and social implications of the rise of AI and the automation economy—what the World Economic Forum calls the 4th Industrial Revolution.
On the bright side, the group highlighted the potential of AI to bring about more equity and access to new jobs, tools, and resources. However, the group expressed concern over the increasing potential for algorithms to exacerbate human bias in who gets a loan, where police patrol, and who goes to jail, among other things. In these public applications, we need transparency around who owns the algorithms and how they work.
Deep discussion centered on the future of work and the threat of widespread automation. Job dislocation, fueled by the effects of AI and robotics, has the potential to exacerbate wealth gaps and thwart upward social mobility. While Scandinavian countries are experimenting with new income protection schemes and China has a plan to dominate in AI, the implications of AI are absent from Washington DC conversations. The group had suggestions ranging from the creation of a federal AI czar to organizing a group of forward-thinking mayors to asking superintendents to articulate the skills and behaviors that schools will need to more explicitly develop in their students.
Given this uncertain future, Russ noted the increasing importance of social emotional intelligence and creative problem solving, as well as more pathways for continual reskilling and upskilling. Curt Allen, Agilix, urged building and using tech with empathy.
Where and how to start the conversation? The group concluded there should be robust conversations locally, regionally and nationally.
How to prepare?
Like the tech group, they agreed that social skills and emotional intelligence are more important than ever.
Students need to be able to think critically about the use of new technology. Computational thinking will be necessary as so much of our lives will be controlled by algorithms. We will be collaborating with tools as well as teammates.
Navigational skills (what NGLC calls wayfinding abilities) are key in the gig economy. Also key, is iterative thinking – the metacognitive ability to revisit assumptions and beliefs to stay current as change is accelerating.
The technologists advocated for developing good science-minded, artistic-minded humans who will take care of the planet.
Marie argued that fundamental to these new priorities is the concept of agency and experiencing success in making contributions to things and people that are important to them. This sense of agency will allow young people to continuously reinvent their skill sets. (Listen to Bjerede explain more in this podcast)
How to develop these new priority outcomes? Inquiry based learning–including project, problem, and placed-based learning–encourages students to ask questions and build solutions. It builds student agency.
Boyd said we should be considering ways to use technology to make us more resilient.
On learning spaces and strategies, the group urged communities to start with what graduates should know and be able to do, THEN design schools/classrooms based on that.
For more on the new economy see:
- Ask About AI: The Future of Learning and Work, a new report from Getting Smart
- Shaping the 4th Industrial Revolution, podcasts and book from World Economic Forum
- The Intention Economy, a book by Doc Searls
For more on education implications see:
- Portrait of a Graduate, Battelle for Kids
- Visioning Toolkit: Laying the Groundwork for a Community-Wide Vision for Personalized Learning, KnowledgeWorks
- MyWays from NGLC