Today’s (10 Sept 2026) Mumbai edition of the Times of India featured two intriguing items concerning Artificial Intelligence. The first involved an AI safety researcher warning that developers in leading labs fear advanced systems could pose an existential threat before the end of the decade. The second reported ambitious claims regarding AI systems tackling major mathematical milestones, such as the Millennium Prize Problems
These two narratives illustrate diametrically opposed visions for humanity’s future. Before addressing them, one must keep in mind that roughly 99.9% of all species that have ever lived on Earth are now extinct. There is no biological reason to assume that humanity is exempt from this rule. Charles Darwin never suggested that evolution culminates with human beings; he viewed humanity not as nature’s ultimate product, but merely as one branch among countless others.
The formal discipline of Artificial Intelligence emerged after British mathematician Alan Turing published his seminal paper, “Computing Machinery and Intelligence,” in 1950. In 1955, John McCarthy coined the term “Artificial Intelligence” in his proposal for the 1956 Dartmouth Summer Research Project. McCarthy also created the Lisp programming language in 1958. Practical robotics followed swiftly: General Motors introduced the first industrial robot, Unimate, to its assembly line in 1961 to handle hazardous die-castings and spot welding. In 1979, the American Association for Artificial Intelligence was founded; it has since been renamed the Association for the Advancement of Artificial Intelligence (AAAI).
Data aggregated by Our World in Data tracks how modern AI benchmarks against human performance across several domains. Between 2010 and 2024, AI systems reached or exceeded human parity in six key capabilities: handwriting recognition, speech recognition, image recognition, reading comprehension, language understanding, and predictive reasoning.
AI has steadily automated routine decision-making across numerous sectors. Algorithms now routinely manage passenger airline ticketing, loan eligibility evaluations, and initial resume screenings. Recommendation engines, such as YouTube’s, dictate viewer feeds based on behavioral history. In healthcare, computer vision aids diagnostic imaging, while robotic platforms assist surgeons with high-precision procedures.
Beyond data processing, machine learning has begun to assist pure mathematics. Rather than merely calculating known values, systems are beginning to assist in formulating novel conjectures. Projects like the Ramanujan Machine uncover algorithmic patterns and propose continued-fraction conjectures for fundamental constants, which human mathematicians subsequently formalize and prove. When AI systems map complex logical spaces, they inevitably highlight unproven edge cases and structural anomalies, generating new avenues for mathematical research.
Monopolies and the Enclosure of AI Research
A major barrier to realizing AI’s potential is the paywall infrastructure maintained by commercial academic publishers. Media theorist Howard Besser observed that just as economic forces drive the gentrification of cities and erode physical public spaces, commercial interests have enclosed digital information spaces. Economists Michele Boldrin and David K. Levine examined this phenomenon in Against Intellectual Monopoly, arguing that broad IP protections frequently hinder innovation rather than foster it.
This issue was central to Google LLC v. Oracle America, Inc., regarding copyright enforcement on Application Programming Interfaces (APIs). In an amici curiae brief, prominent computer scientists emphasized that open interfaces are fundamental to computational progress:
“Excluding APIs from copyright protection has been essential to the development of modern computers and the Internet. For example, the widespread availability of diverse, cheap, and customizable personal computers owes its existence to the lack of copyright on the specification for IBM’s Basic Input/Output System (BIOS) for the PC.”
If knowledge silos and aggressive IP claims restrict access to training architectures, development will consolidate within a handful of corporate monopolies.
Evolution on Earth began approximately 3.8 billion years ago, progressing across biological substrates and now extending into technological ones. While alarms over imminent extinction may often carry rhetorical hyperbole, ungoverned deployment of autonomous systems remains a genuine risk. Isaac Asimov’s classic “Three Laws of Robotics” offer a foundational philosophical reference for safety and human alignment:
- The First Law: A robot may not injure a human being or, through inaction, allow a human being to come to harm.
- The Second Law: A robot must obey the orders given it by human beings, except where such orders would conflict with the First Law.
- The Third Law: A robot must protect its own existence, as long as such protection does not conflict with the First or Second Law.
To reflect on the twilight of our biological primacy, Arthur C. Clarke’s Childhood’s End provides a haunting portrait of species-level transition:….
“There was nothing left of Earth. They had gone over it like a light that illuminates and consumes. They had left nothing behind them, and they were gone.
…
Jan Rodricks stood alone upon the dying planet… Earth was no longer his home, nor the home of man. It was only the launching pad from which humanity had leapt into the stars.”
References:
Boldrin , Michele and Levine David K , Does Intellectual Monopoly Help Innovation? In Review of Law & Economics, 2009
Howard Besser, Intellectual Property: The Attack on Public Space in Cyberspace, http://www.gseis.ucla.edu/~howard/Copyright
Julie P. Samuels, Attorneys for Amici Curiae Computer Scientists, Brief of Amici Curiae Computer Scientists in support of Defendant –Cross Appellant and Affirmance