Governance, Ethics, Cybersecurity: what we learnt in Phase 2 of the Global Debate
As we advance the discussions on the AI Civic Debate, here is a short synthesis to help you navigate the insights gained so far. The discussion threads have already revealed the salient features in governing AI and suggested preliminary concrete solutions.
We thank you so much for your proposals and ideas so far! The process of collective intelligence is leading us to deepen your proposals and ideas in Phase 3. You will see your ideas discussed in greater depth in the coming weeks.
We cant’ wait to launch Phase 3, but until then, enjoy the synthesis!
Reinventing Man & Machine Relationship
The discussion has highlighted the importance of the AI revolution: whether ‘weak’ or ‘strong’, AI will transform our lives. Embedding ethics and values in machines makes humans reflect deeply on our own values and the notion of “emotion” as well “relationship”. One participant took the example of the movie “Her” to show that we will have more and more deep-bonding relationships with machines. Participants have also debated whether AI would help us fight our own biases, counteracting deep seated issues of algorithmic biases. What do you think?
“Although the strong A.I. would not emerge in this century, the weak A.I. could transform the day life of our human beings and the economic system of capitalism, and the meaning of humanity.” KeeYoung Yoon
“I’m concerned that merely transposing existing bias into machine learning products will not replicate, but pro-actively scale bias. To me, it seems the existing choice we have is to scale or to stop existing biases.” Marwan
“we should create ISO standards for AI/human rights and algorithm audits for those standards.” Jerome C. Glenn
AI Safety & Security
Participants highlighted the increasingly deferred control of decisions to AI could lead to an loss of power over machines and risk longer term human safety. The community also argued that having access to more data and better recognition systems could solve cyber-security issues.
“I agree that allowing a greater “flow of data” will help limit A.I.’s vulnerabilities to false data, in addition to the equally important implementation of “critical thinking” skills.” Ria
“IMHO broadbrush approaches are needed in the short term – Eg. improve the checks and balances in policy making to push those with political power to better consider future generations and global catastrophic impacts” Samuel H.
The question of an arm race was also debated. How to avoid it? The United Nations was evoked as the most widely accepted international body to take some action on forbidding automated weapons and AI arms race, but was recognized as slow to move and maybe ineffective in the short term. Do you agree or disagree?
Governing AI is a debate nested in privacy, whether we should allow a free flow of data, which improves AI accuracy, or strengthen boundaries between personal and other data. Some participants side with right to privacy is reason enough to limit free flow of data, while others assessed that more information could lead to a more honest, and secure, society. Participants had mixed answers regarding regulations, but in all cases, policymakers and large companies should pay more attention to public opinion, who feel left out from big decisions.
“I would say the best strategy to make the biggest impact would be to (if money allows) create a corporate entity that is made up of both a lobby group and a regulations group. Lobby congress for funds to do research/boards into AI regulations and public outreach. Get AI business (IBM, Facebook, Amazon..) to sign up for self-regulation and help fund the public outreach.” Matthew R.
Adapting the workforce for the Age of AI
This is arguably the most important topic among our community. Participants quickly agreed that it was central to decide what skills humans should acquire in the age of automation. One participant proposed to make learning coding compulsory, while others argued that we should focus on more creative and social skills that machines won’t be able replicate, since it is the essence of being “human”. Given the rapid wave of automation coming in the next few years, the conversation quickly moved to the creation of a Universal Basic Income. Other creative ideas were shared, as well regulation of automation to prevent job losses. What are your views?
“If we want to be able to regulate and control our relationship with the AI, we must be able to understand how it works and be able to intervene in the event of contingencies, and to acquire this expertise, it is better to start very early.” Almamy A. Z.
“Countries coordinate & agree on an income distribution or other political economy model for the future – where they share wealth – before free-market forces naturally determine it.” Yolanda L
Drive AI for Public Good
How can AI benefit everyone? Participants highlighted the importance of providing free services that will enable greater access, providing services such as access to legal tools to make our communities more inclusive. Making high quality knowledge open source and accessible to as many people as possible was highlighted as an important criteria in order to include various demographics in society on topics in AI. The idea of “Creative Commons” was proposed as a concrete way to empower individuals in making choices.
“The idea is that the power of the AI does not remain the prerogative of a few large specialist groups, consulting firms, TNCs, but that everyone benefits from these unsuspected advantages.” – Cédric G.
“The role of governments is to protect the human citizens against this new tool in the hands of companies. The roles of companies is to use this tool to help humans” tenten
Imaginaries of AI
As we invite our community to write a fictional story that reshuffles the way AI will impact our everyday life, we have experienced great creativity! Participants bring ideas from the invention of an “AI-brid” sentience, to “the death of GDP”and the rise of the “Super Intelligence Index”. AI seems enable self augmentation, interstellar travels and inner exploration… We are excited for new short stories to emerge – Share your imaginary!
“January 15, 2050. The night has fallen. I was settling for another mind blowing evening. Forget what was called “TV” still two decades ago. Boring! What now seems to become a routine to me, would have still sounded impossible 6 months ago. But nothing was surprising anymore, with Moore law passing the vertical exponential tangent of progress in all aspects of “emerging” (emerged, really) technologies….” Cyrus
“January 2050, I was reading quietly in my living room, watching my son, a 6 year old, making his own drone from an open source website he found by himself” Mohamed Zayed
Thank you for your participation!
Join us for Phase 3, starting February 1st!