Publish at May 17 2023 Updated May 17 2023
"As a precaution, we make extremely heavy machines with very small legs."
Larry Page
A hundred luminaries, including a certain Bill Gates, are calling for a moratorium, probably because the consequences of artificial intelligences are currently poorly assessed, and inconsequential decision making is contemplated with ethical, societal, but also safety risks, which is sobering when you know the possible devastation of nuclear weapons. AIs challenge head-on what it means to be human. The effects to follow point to the breadth of questioning at work by philosophers, thinkers, engineers, and artificial intelligence experts.
One of the most fundamental philosophical questions posed by artificial intelligence is whether a machine can be considered intelligent without understanding what it is doing. According to John Searle,
"What we want to know is whether there can be artificial intelligence without understanding, and whether such intelligence can be autonomous, and if not, why not" (1980).
If machines can be considered intelligent, this also raises questions about the relationship between intelligence and consciousness. Western consciousness had been turned upside down after the discovery of America, when will it be after the emergence of the hidden continent of AI?
Nick Bostrom suggests that
"The very concept of a nonbiological intelligence requires a radical revision of our ideas about who we are and what our place is in the universe"(2014).
This shift in the link between consciousness and intelligence challenges the foundation of our humanity, not all of the implications of this dissociation are clearly drawn to date.
Artificial intelligence may have important implications for anthropology with respect to our understanding of human beings and our relationship to technologies. According to Yuval Harari,
"The advent of intelligent machines threatens to disrupt human societies by making many jobs obsolete and creating new forms of social and economic inequality" (2018).
According to a Oxford study, machines will outperform humans in a wide range of tasks by 2060. Published in 2017, the study explained that computer programs would be able to write entire books as early as 2050, or perform certain surgeon tasks by 2053. According to Schumpeter's principle of creative destruction, an innovation comes to destroy old trades and practices to make room for others that carry greater efficiency, however in history, those left behind in the old practices have suffered declines, what political risk do we run in letting the market adjust according to its "natural laws"?
Similarly, Kate Crawford and Ryan Calo have pointed out that
"The development of AI raises fundamental questions about what it means to be human and how we relate to machines that mimic or surpass human intelligence" (2016).
Since the ancient myth of the Golem, we have been constantly comparing ourselves to the machines we create even though we are not able to compete with them on a range of computational but now also analytical and prospective functions. We used to think that reserved fields such as creativity would hold out until with a few clicks incredible images appeared with Midjourney.
The ethical concerns related to artificial intelligence are varied, ranging from the risks of discrimination and bias to the question of accountability and control. According to Joy Buolamwini and Timnit Gebru,
"AI is only as unbiased as the data and algorithms that drive it" (2018).
Or, a multiplicity of biases in the design of algorithms is detectable and they end up reproducing the majority discriminations and beliefs of the moment since AIs rely on consulting a mass of data. Ethnic minorities, women, and those left behind leave with a handicap if nothing comes to correct the mass of already biased data on which the calculations are based.
The ethical challenges of AI also include transparency, accountability, and control, as Stuart Russell and Peter Norvig point out:
"The ethical challenges of AI include not only the risk of discrimination and bias, but also issues such as accountability, transparency, and control" (2020).
A Stanford survey notes "AI-related 'incidents and controversies' have increased 26-fold over the past decade. Cases of addiction, suicide, divorce have even been blamed on conversational agents.
The existence of artificial intelligence raises profound questions about the nature of reality and how we relate to intelligent machines. According to Luciano Floridi,
"AI is a new form of life that challenges our traditional ontological categories and forces us to rethink our relationship with the world" (2014).
The question of whether machines can be endowed with artificial consciousness is also relevant, as David Chalmers has pointed out:
"The existence of intelligent machines raises profound philosophical questions about the nature of reality and the possibility of artificial consciousness" (2010).
If AIs do not yet dream, have little humor, the interconnection of AI functions will increase suspicion of a generalized society of control. The crossing between databases, facial recognition and automatic decision making is questionable. This is already underway on the Asian continent, but interconnected systems are already in place, for example in the coupling of radar to control the speed of vehicles, addressing and collection of fines. Little by little we are getting used in a variety of daily acts to this mechanization of the world.
The social implications are already significant, especially with regard to employment and security. According to Erik Brynjolfsson and Andrew McAfee, "Smart machines may replace human workers in many jobs, creating economic and social disruption" (2017), because while new jobs may well be created to design and drive AIs, old ones are set to disappear abruptly. In addition, concerns about safety and privacy are significant, as Edward Felten points out:
"AI can create new safety and privacy risks, and we need to be aware of these risks so we can manage them responsibly" (2019).
AI fraud is growing, the production of false information and evidence are becoming plagues on democracies and distilling the poison of doubt. Finally, the development and application of artificial intelligence also raises significant technical challenges, particularly in relation to resource consumption. According to Yann LeCun, Yoshua Bengio, and Geoffrey Hinton,
"One of the key challenges in AI is to develop algorithms that can learn from data efficiently and robustly" (2015).
In doing so, not only do AIs produce infobesity, they plunder any intellectual property by feeding on data in a legal limbo. Moreover, AI consumes significant computing and energy resources, and raises questions about digital's carbon impact projected to triple by 2050.
All of these issues escape the fine-grained understanding of citizens and individuals. A few experts and companies are committing us without consultation to a path with major consequences, without counter power, which violates a major democratic principle.
Sources
John Searle (1980). "Minds, brains, and programs." Behavioral and Brain Sciences, 3(3), 417-424. www.course.sdu.edu.cn/G2S/eWebEditor/uploadfile/20140227112825015.pdf
Nick Bostrom (2014). Superintelligence: Paths, Dangers, Strategies. Oxford: Oxford University Press.
https://www.decitre.fr/livres/superintelligence-9780199678112.html
Yuval Harari (2018). 21 Lessons for the 21st Century. New York: Spiegel & Grau.
https://www.decitre.fr/ebooks/21-lessons-for-the-21st-century-9781473554719_9781473554719_10012.html
Kate Crawford and Ryan Calo (2016). "There is a blind spot in AI research." Nature, 538(7625), 311-313. doi:10.1038/538311a
Joy Buolamwini and Timnit Gebru (2018). "Gender shades: Intersectional accuracy disparities in commercial gender classification." Proceedings of the 1st Conference on Fairness, Accountability and Transparency, 77-91. doi:10.1145/3176349.3176356
Stuart Russell and Peter Norvig (2020). Artificial Intelligence: A Modern Approach. Upper Saddle River, NJ: Pearson Education. https://www.decitre.fr/livres/artificial-intelligence-9781292401133.html
Luciano Floridi (2014). The Fourth Revolution: How the Infosphere is Reshaping Human Reality. Oxford: Oxford University Press.
David Chalmers (2010). "The Singularity: A Philosophical Analysis." Journal of Consciousness Studies, 17(9-10), 7-65.
Time Elon Musk and 100 experts call for a moratorium on artificial intelligence
https://www.letemps.ch/economie/cyber/elon-musk-experts-reclament-un-moratoire-lintelligence-artificielle
France24 Artificial Intelligence: the jobs that will disappear and those that will replace them
https://www.france24.com/en/emissions/tech-24/20230113-artificial-intelligence-the-jobs-that-will-disappear-those-that-will-replace-them
France24 Democracy: deepfake in South Korea a fake candidate to help the real one
https://www.france24.com/fr/info-en-continu/20220214-la-démocratie-deepfake-en-corée-du-sud-un-faux-candidat-pour-aider-le-vrai
Futura sciences. Global warming: the digital carbon footprint is exploding because of AI
https://www.futura-sciences.com/planete/actualites/rechauffement-climatique-empreinte-carbone-numerique-train-exploser-cause-ia-104727/
MSM Technology and Science. Is AI leading us to a nuclear level disaster
https://www.msn.com/fr-fr/actualite/technologie-et-sciences/l-ia-nous-mène-t-elle-vers-une-catastrophe-de-niveau-nucléaire/ar-AA1a2uDK
You might also like:
Trust, the basis for living well in society, yes, but which society are we talking about?
With COVID, trust in one's co-citizen has fallen to an all-time low. It can be contaminated and therefore it can be dangerous. A devious phenomenon because it is intangible, it affects our lives. A phenomenon known from armed conflicts, it has arrived through this biological war. What are the social effects? Can we regain lost confidence? Can we move forward in a society of presumption?
Adding value to education with comprehensive education
Educational value has often been based on mastering school-related skills, i.e. reading, counting, writing, etc. However, there are more inclusive pedagogies that seek to train students in their bodies and minds as well as in their hearts. This approach is highly desirable but not easy to put in place.
Organization, management and skills: the new deal
Which organization, which key competencies and which management to rebuild after the crisis? The organizations that were presented as examples of success a few months ago are no longer so, employees' aspirations are different and teams are dispersed when it is necessary to work as a group... Many employees have made a thirty-year technological leap in two months!
Three forms of contemporary influence
In the Theater of Influencers, I introduce you to the new players on stage: the facilitator, the algorithm and the social influencer.
How much freedom does excitement give us?
So that we don't just suffer the events of our lives. We often can't change events, but we can change the way we react to them.
