The Limits of "Techno-Solutionism"

It is undeniable that technology, when harnessed for human creativity, has brought immeasurable benefits to our lives and paved the way for a better future in many regions of the world. However, the omnipresence of technology and its impact on the lives of billions, connecting everything everywhere and at all times, now raises some crucial questions: What is the place of humanity in a world dominated by technology? What directions do we want progress to take, with what goals and means? In his 2014 book To Save Everything, Click Here, Evgeny Morozov offers a clear definition of what technological solutionism is. For him, the development of new technologies has been accompanied by the rise of “two ideologies.” The first, "solutionism," is a totem of the Silicon Valley evangelists, who believe that all aspects of our lives can be improved and most problems solved through and by technology. The second, "web-centrism"—a new fantasy driving the bloated economy of digital transformation—has made the digital realm the center of the world's gravity. The rapid digitization observed over the years is not without consequence. We have seen countless abuses: increasing misinformation, leaks, and illegal exploitation of user data, phishing scams, identity theft… Where once platforms were seen as the new realms of contemporary utopias, they have, over time, fostered the emergence of voracious individualism and unashamed egocentrism, far removed from the spirit of the internet pioneers, born out of the libertarian counterculture of the 1970s (1).

Far from the infighting and the irreconcilable duality between technophiles and technophobes—whose opposition is likely too binary and schematic to be fair—I have observed for years how new technologies shape our lives more each day and lay claim to the long term.

Today, everything seems to be moving fast, too fast, without this alarming our political leaders, without any real reflection on the social, environmental, and ethical consequences. Over time, technological innovation has supplanted the notion of progress; worse, it increasingly appears to be its only manifestation. This reality should enlighten us to the magnitude of the task at hand and stimulate our often resigned intellect. Through this article, I will attempt to answer a question many are asking: What if technology creates more problems than it promises to solve?

Rare Earths, An Endless Race

The frantic quest for precious rare earths—the new black gold—necessary for the proper functioning of our smartphones, computers, and electric cars is degrading terrestrial and marine spaces. From lithium extraction in the salt desert of Atacama in Chile, to cobalt, tungsten, and coltan mining in Inner Mongolia and the DRC, to polymetallic nodule extraction from ocean floors, humanity is destroying ecosystems daily, on land, along coasts, and in the seas, to satisfy its technological hunger. Because nothing seems to fulfill the endless aspirations of a humanity that has lost its compass, the major powers are now engaging in a real race to extract exploitable metals from the Moon, particularly helium-3, considered the energy of the future...

Supposed to decarbonize the economy through the energy transition, the green revolution seems increasingly distant and is running into certain realities. For producing many “renewable” energies, rare metals are needed.

Blaming wind, solar, hydro, or biomass energy for the world’s pollution due to mining activities is sophistic. However, we must ask ourselves whether the remedy is worse than the disease and thus question this outrageously fantasized "green revolution." In The Dark Side of Green Energies, Guillaume Pitron clearly shows that every so-called “green” technological innovation inevitably causes pollution transfers. For Aurore Stephant, a mining geological engineer, it is imperative to urgently rethink approaches to digital transition and identify possible levers to drastically reduce associated impacts.

Projects consuming rare earths are multiplying everywhere. "Spot," Boston Dynamics' robot dog, now secures nuclear sites, participates in first aid missions, and intervenes in fires. Homo Numericus Elon Musk, CEO of Tesla and SpaceX, aims to produce millions of units of his new humanoid robot "Optimus" to "build a future of abundance and fundamentally transform civilization."

In the Cloud—now the holy grail of digital giants—lie countless servers, terminals, rare earths, and waste. The CO2 consumption needed for the development of artificial intelligence algorithms—just to mention them—leads to a disastrous carbon footprint. According to researchers at the University of Massachusetts Amherst, training neural networks (a method of artificial intelligence also known as deep learning) can consume as much energy as five cars, from their production to the end of their life. The study was noted by the MIT Technology Review. Finally, while the issue of water usage in data centers is a closely guarded secret, the Netherlands—hit by drought—recently discovered the staggering consumption of Microsoft’s data center: 84 million liters of water, while the American company had indicated that its facility only needed 12 to 20 million liters... Currently, CO2 emissions and the technological innovation of carbon capture and storage (CCS) are at the center of all discussions. While some consider carbon credits a pragmatic solution to the planet's climate problems, others argue that they exacerbate the issue, giving polluters the freedom to emit more than they otherwise would. Although this solution should not be definitively dismissed for now, carbon credits and offsets associated with blockchain and cryptocurrencies have never demonstrated their full effectiveness and ability to reduce greenhouse gas emissions in a clear and sustainable way.

Moving Beyond Moore's Law

The endless "technological race"—sold as a categorical imperative—is at the heart of geopolitical issues and the Sino-American rivalry. While 5G is just beginning to appear, Chinese satellites are already testing 6G... Aware of the strategic applications in military and civilian domains, states are already investing billions in the sixth generation of wireless communication technologies. According to LG, 6G could be commercialized as early as 2029, though it still has several technical hurdles to overcome in fundamental research. Meanwhile, American companies like Intel, Qualcomm, Google, and Cisco; Asian companies like Huawei, LG, NTT Docomo, ZTE, and Samsung; and European companies like Nokia, Ericsson, Atos, Orange, Siemens, and Telefonica (project Hexa-X) are already sharpening their weapons to win the 6G battle.

"In an extremely mobile society where everything is disposable, the psychological need for security and stability is often perversely exploited by the manufacturer, the advertising agency, and the merchant, who divert the consumer's interest towards the superficial ornaments of a fleeting 'in-group.'" - Victor Papanek

Like 5G and all the generations that preceded it, devices need to be in constant evolution. The compatibility of devices related to their design is one of the major challenges for manufacturers and at the same time the main problem. Moore's law, which states that computing power more than doubles every two years, has led to the emergence and mass adoption of smaller, more practical, and more efficient devices. While there has been a significant slowdown in the prophetic and empirical logic of Moore's Law—due to the limits of component miniaturization—many players, such as IBM and Samsung, want to push its boundaries, particularly by manufacturing vertical transistors that have so far been constructed horizontally.

This logic only increases the desire for objects of subordination and dependence, which—ironically—were supposed to make us freer... This constant drive for innovation, conducive to planned obsolescence, pushes us to consume more and more, making the fight against digital consumerism futile and illusory.

The book Thrift and Thriving in America reminds us of the importance and benefits of "frugality" and how this moral value articulated the normative dimensions of economic life during the 17th and 19th centuries and throughout much of American history.

A Technological Solution to Every Problem

Instead of addressing the root causes of problems, states and other institutions reflexively turn to technological solutionism. This is the case with managing migration flows through the use of new identification and surveillance technologies that allow for tracking and tracing individuals. Although useful and sometimes necessary, these systems—integrating chips into visas and residency documents, chip readers, cameras, and biometric identifiers linked to databases—are part of a reactive system and not a proactive policy aimed at anticipating migration flows upstream.

"All science, having become experimental, depends on technology, which alone allows for the technical reproduction of phenomena. However, this technology abstracts nature to allow for scientific experimentation: hence the temptation to force nature to conform to theoretical models, to reduce nature to technoscientific artifice." - Jacques Ellul

Technological progress has become a formidable alibi to absolve us of our missteps and the tool for correcting our greatest follies.

By clumsily addressing the root of the problem, we end up responding to it in the worst possible way. While we now know that the growing use of air conditioning exacerbates global warming, Dubai has invested billions of dollars in gigantic towers and other high-end energy-consuming residences in air conditioning, only to then invest considerable sums in developing technologies that stimulate clouds to make it rain. Climate manipulation through geoengineering could have a catastrophic impact on the atmosphere and ocean balance. At the very least, the relevance of these anthropogenic interventions should be questioned. While some point out the environmental benefits of new practices and the rise of new tools that reduce certain travel, teleworking is one of many examples.

The Metaverse has quickly become the new avatar of a dystopian futuristic world where digital twins promise to synchronize the physical and virtual. Facebook promises surgeons the ability to practice as much as necessary in the Metaverse before operating on real patients and students the chance to listen to Mark Antony debate in ancient Rome thirty-two years before our era.

Located in the middle of the Pacific, the Tuvalu Islands, and their 12,000 inhabitants, could be submerged due to climate change. The state of Tuvalu is considering creating a virtual version of the Polynesian archipelago before it is submerged by the end of the century due to global warming, reports The Guardian.

While the applications are real, Raja Koduri, an executive at Intel, asserts that creating a metaverse for hundreds of millions of users would require computing power a thousand times greater than it is today. Not to mention the renewal of devices necessary for our new immersive life: from phones to virtual reality headsets and glasses, to computers (2). According to Fabrice Flipo, "it is crucial to quickly establish a market authorization system that would force companies to produce impact studies on the socio-ecological trajectory of their digital projects." (3)

The Age of Surveillance Capitalism

The digital age has brought many advances and with them a host of disillusions and new threats to our privacy. One of the greatest dangers lies in how big tech companies collect and use our data.

According to academic Sarah Spiekermann, "Some companies hold up to 30,000 data points for each individual they track."

Originally, the internet was designed to be a decentralized network, where each user could connect with any other user without passing through a central server. This design was based on the belief that decentralization would make the internet more resistant to censorship. The various personal data regulators—CNIL, the California Consumer Privacy Act (CCPA), the European GDPR, or the Health Insurance Portability and Accountability Act (HIPAA)—constitute firewalls but unfortunately do not prevent abuses. The collection and transmission of the personal data of millions of French people to the giant IQVIA—the world leader in the collection and analysis of medical data—is proof of this. The American company now holds, with the consent of the CNIL and many pharmacies, all kinds of medical information, including pharmaceutical prescriptions. Although the data is "pseudonymized" to make re-identification of individuals as difficult as possible, the operation is reversible, unlike anonymization.

The Omnipresence of Algorithms

A source of progress in many areas and scandals of all kinds, algorithms fascinate as much as they frighten. Now, they dictate the social lives of individuals and too often fuel divisions and fractures. As such, we must question their various applications. In her book The Age of Surveillance Capitalism, Harvard emeritus professor and sociologist Shoshana Zuboff denounces the "monstrous" power of intelligent machines, a power that depends less on their present interactions than on their ability to predict and shape our future behaviors. Without denying their real contribution to all aspects of our lives, artificial intelligence is clearly under scrutiny. By automating existing fractures and discriminations, digital transformation only reinforces them. Scientists from New York University's AI Now Institute have recently raised concerns about the lack of regulation around AI. The fears raised in their report particularly concern facial recognition technologies and algorithmic biases: "It is increasingly clear that in various fields, AI amplifies inequalities, placing information and control in the hands of those who have power," the document reads.

Artificial Intelligence in Healthcare

Artificial intelligence, particularly machine learning, can greatly assist in processing medical data and provide healthcare professionals with important information, thereby improving health outcomes and patient experiences. Structuring data from patients, facilitating diagnosis, guiding therapeutic care, and aiding in prediction in oncology or cardiovascular fields—the examples of artificial intelligence applications in healthcare are numerous and undeniable. Although AI offers significant guarantees in terms of security, it remains a vulnerable technology, particularly sensitive to so-called "adversarial" attacks and manipulations that can alter AI behavior. By changing a few pixels on a lung scan, it is possible to deceive AI into seeing a disease that does not exist or not seeing one that does, explains Samuel Finlayson, a researcher at Harvard Medical School and MIT, and author of the article Adversarial Attacks on Medical Machine Learning.

The article emphasizes a growing sense of concern about the possibility of adversarial attacks that could affect facial recognition services, autonomous vehicle systems, or even iris scanners and fingerprint readers. In 2018, a team from NYU's Tandon School of Engineering succeeded in creating virtual fingerprints capable of fooling fingerprint readers 22% of the time. In other words, 22% of all phones or PCs using these readers could potentially be unlocked...

Given such findings, we must question the path we are taking and whether it is desirable to use algorithms for new applications that may one day surpass us. Timnit Gebru, one of the leading researchers in the United States according to Nature magazine, was dismissed by her employer Google for criticizing the company's minority recruitment policies and the biases of its artificial intelligence systems. In her book Weapons of Math Destruction, Cathy O'Neil, a Ph.D. in mathematics, a Harvard graduate, and an algebraic geometry researcher, strives to demonstrate how and why "Big Data increases inequalities and threatens democracy."

Scientists from the AI Now Institute at New York University have recently raised concerns about the lack of regulation around AI.

The concerns raised in their report particularly relate to facial recognition technologies and algorithmic biases: "It is increasingly clear that in various fields, AI amplifies inequalities, placing information and control in the hands of those who have power, thereby further reducing those who already have little," the document reads.

Social Credit, The World to Come...

Often described as the new black gold of the 21st century, personal data—coveted by Chinese and American giants—represent precious and strategic resources for large companies and states.

Although they deny it, many countries no longer hesitate to praise the merits of the Chinese counter-model and even go so far as to question the effectiveness of the methods used by democracies. Countries like Sri Lanka, Cambodia, Chile, or Poland express interest in China's social credit system. In France, some politicians, including Senator Jean-Raymond Hugon, advocate for the implementation of a French social credit system. The idea is gaining traction in France and some European countries. While the French model bears no resemblance to the Chinese one, the vaccine passport follows the same logic that justified the Chinese social credit system. Caught between the temptation of Chinese social credit and the arbitrary censorship of American platforms that ban any contradictory debate, our democracies appear fragile. Without denying the need to provide appropriate responses to an exceptional situation and beyond the troubling question of the habit of social control, what remains of the legitimacy of states if each crisis, whether health or security-related, highlights their inability to cope without resorting to solutions that lead to greater coercion? In times of turmoil, democratic instability—becoming more frequent across the West—technological frenzy cannot be the only solution.

No one knows if one day, under the pretext of better taming creeping anomie, the vaccine passport—just to name one—will be used to generalize social control in France and further weaken interpretative freedom.

The Urgent Need for Regulation

The availability and exploitation of health data have become one of the major issues of our century. These so-called sensitive data are subject to special protection conditions regulated by the CNIL and GDPR. Yet, since 2018, the Commission Nationale de l'Informatique et des Libertés (CNIL) has authorized more than 140,000 pharmacists to collect the pseudonymized personal data of millions of French citizens, which are then transmitted to the largest American data broker, IQVIA. While the American giant claims not to use this "anonymized" health data for commercial purposes, the risks are very real. Tech giants, as well as major insurers, are investing considerable sums in this health sector, and partnership initiatives are multiplying, such as the agreement signed between the UK National Health Service (NHS) and Alphabet, which involved processing data from 1.6 million patients. Doubts remain about the guarantees of security, integrity, and confidentiality in the collection and processing of patient data!

In 2018, the company Grindr was at the center of a major scandal. The app had shared the personal data of its subscribers, including their HIV status, with companies.

In recent years, predictive Big Data analytics and processing technologies have been multiplying everywhere. Palantir Technologies, specializing in data exploitation for decision-making and prediction, is probably the most well-known. Funded by In-Q-Tel, the CIA's venture capital arm, this data exploitation company was used by the New Orleans police in 2012 to predict which individuals were likely to commit crimes. (4) While technology can undoubtedly help law enforcement fight crime (digitization and 3D printing of crime scenes, morphological predictions, criminal act anticipation...), the New Orleans example shows the limitations of predictive technologies, which highlighted numerous errors and major biases, reinforcing social, racial, and economic prejudices in this particular case. This example forms the basis of the famous Lucas Critique theorized by Nobel laureate Robert E. Lucas.

Another example is price discrimination—based on browsing history recorded by cookies—introduced by some companies. Nowadays, some internet users know how to manipulate the prices of tickets offered by airlines by changing their behavior. It is thus quite easy for them to fool the cookies responsible for this discrimination to obtain better deals. Following this example, it is easy to understand why ex-ante control by any authority of the tools provided by artificial intelligence would be illusory. However, concrete applications do exist. Researchers and doctors are experimenting with virtuous applications of artificial intelligence. At Bordeaux University Hospital, the goal is to optimize patient care and save time for caregivers.

"In the field of health, ethical progress aims to repair humanity, non-ethical progress aims to augment humanity." - Axel Kahn

The question of better regulation is more pressing than ever. In Europe, the Digital Services Act (DSA) and the Digital Market Act (DMA) aim to establish a set of common rules and regulate the European digital space. In the United States, the Algorithmic Accountability Act was introduced in the Senate and House of Representatives on February 3, 2022. China has also drastically regulated the use of recommendation algorithms in recent years, particularly with the implementation of the PIPL ("Personal Information Protection Law"), China's comprehensive personal data protection law modeled after the European GDPR.

The European AI strategy presented in 2021, aimed at making the EU a world-class hub for Artificial Intelligence, seems to be heading in the right direction. This was the focus of my work—sent to Margrethe Vestager—and for which I devoted an article in 2018. However, the funds allocated to support such ambition are woefully insufficient. Under the Digital Europe and Horizon Europe programs, the Commission plans to invest 20 billion euros over the digital decade. By comparison, the Chinese and Americans plan to spend three to four times more when public and private investments are combined.

Digital Pharmakon

The new technological paradigm symbolized by facial recognition, the rise of the Internet of Things, mass data surveillance, hyper-automation, the tyranny of algorithmic biases, geolocation technologies, and the proliferation of surveillance cameras brings our societies closer each day to the dystopia of George Orwell's 1984.

While we are not yet in the world of Isaac Asimov and John W. Campbell's Robot Series, pollinator drones, robot dogs patrolling borders, and humanoid robotics with ultra-realistic facial expressions—the latest avatars of our technological obsessions—bring us inexorably closer. This technological race, widely denounced by Jill Lepore, professor of history at Harvard, should not make us forget that, unlike progress—which seeks to improve the human condition—innovation is content to create and launch new products, leaving it to lawmakers to address ethical issues.

"The technology already exists. It’s only the will we’re lacking." - Cathy O'Neil

Whether it involves supporting the needs of the elderly and those with reduced autonomy, addressing the challenge of seawater and brackish water desalination, reducing maternal and infant mortality, or tackling the great challenges related to genetic diseases and cancers, the needs are numerous, and the challenges are enormous. Thus, genome editing technology—CRISPR—functions as a programmable molecular scalpel and could eventually allow the treatment of the most severe genetic diseases. The Neuralink brain implant, promising to help people with disabilities, opens up new possibilities, particularly by offering the potential to improve or cure conditions such as Parkinson's and Alzheimer's. Unless these technologies, on the contrary, open Pandora's box to the worst of transhumanism. For while technology is neutral, the ethics of the user and the designer are not always so. Far from wanting to shield ourselves from the lightning of progress, we must accompany this great transformation by putting safeguards in place. In 2021, the company Clearview AI, which caused a real scandal by collecting nearly billions of photos without any permission, allowed American FBI investigators to identify and arrest an Argentine pedophile who had been on the run for years. Proof that technology is capable of both the worst and the best.

Between the technosolutionists who have made technology an ideology and Alain Damasio's "Furtives," moving in the blind spots of our society to escape the hell of chips and robots, there is probably a path through the nettles and a small light of hope at the end of the cave. The Low Tech movement—which encourages eco-design, resilience, sobriety, and responsibility—localism, the growing questioning of consumption by millions of citizens worldwide, supply chain decarbonization, and the concept of de-digitization of certain sectors are gaining ground. All is not lost if we separate the wheat from the chaff and if all these technological initiatives are passed through the sieve of reason and common sense. Their uses, linked to individuals' intimacy, freedom, and privacy, challenge the very foundations of our democratic societies. The perpetration of the worst and its acceptance will depend on our choices—or lack thereof.

To those who, after reading this article, still wonder whether technology, in its broadest sense, is the answer or the problem, like Bernard Stiegler, I believe it is a pharmakon: both the poison and the cure.

References:

(1) Social Media, Mirror of Our Society | LinkedIn

(2) Fabrice Flipo, Nature and Politics. Contribution to an Anthropology of Modernity and Globalization (openedition.org)

(3) Crime Prediction Technologies: Palantir Secretly Monitored New Orleans Citizens for 6 Years (tv5monde.com)

4o