Page 25 - The Indian EYE 092526

NORTH AMERICAN Newsline SEPTEMBER 25, 2026 | The Indian Eye 25

AI’s Darker Side Poses An Existential

Question To Humans – Act Now

I for Bad: The Darker Side of
Artificial Intelligence turns out
Ato be a landmark book that
stands out for its intellectual rigor,
balance and commitment to evidence
in the heated debate over artificial in-
telligence.
Professor N K Goyal and Dr.
Govind Pathak, the authors, argue
that Artificial Intelligence is no lon-
ger a futuristic abstraction but a gi-
gantic force shaping our economies,
our relationships, and even our sense
of self. The book’s foreword is writ-
ten by Ravi Sharma, President, IIT
Alumni Council.
The book refuses to take the
easy route of either hype or hyste-
ria. Instead, it delivers a challenge:
the harms of AI are now too specific,
too well-documented, and too conse-
quential to be dismissed as mere side
effects of progress, inviting readers
to engage in a more nuanced and
structured public conversation. conditions. what kind of AI to build, and what ing—Black-box systems must be
4. Job displacement—AI is not elim- AI owes society in return. opened to scrutiny. If a system can’t
The Evidence Is In: Harms Are Real, inating jobs wholesale, but it is The Core Question: Who De- be explained, it shouldn’t be making
Not Hypothetical erasing the entry-level jobs that cides, Who Benefits, Who Pays? consequential decisions about peo-
With a methodical and transpar- once allowed young workers to The defining question is not what ple’s lives.
ent approach, the book organizes climb the career ladder, threaten- AI can do, but what it should do, for Economic justice—AI-gener-
the AI debate into six domains of ing long-term economic mobility. whom and at whose expense. The ated wealth must be more broadly
harm—criminal misuse, algorithmic 5. Accountability void—When harm evidence shows that harms concen- shared and AI must help entry-level
dehumanization, psychological de- occurs, responsibility is diffused trate on the most vulnerable: chil- job aspirants to create an enterprise.
pendence, job displacement, account- across developers, deployers, and dren, isolated individuals, entry-lev- Global perspective—AI’s harms
ability voids, and the economics of the users, leaving AI largely without el workers, primarily those with the and benefits are not distributed
AI boom, based on peer-reviewed accountability, so the human cost least power to shape how AI is de- evenly across regions or languages of
research, incident databases, litiga- becomes an unavoidable conse- ployed. Meanwhile, the benefits and the world. Policy must reflect this re-
tion records and regulatory filings. quence. wealth generated by AI risk being gional diversity and socio-economic
1. Criminal misuse—AI is powering 6. Unsustainable investment—With captured by a small number of insti- deviations and refrain from imposing
deepfakes, scams, and extremist more than 5 trillion dollars in pro- tutions and investors, a concentra- a universal template.
content at a scale and speed pre- jected infrastructure buildout, AI tion that may prove unsustainable
viously unimaginable. Tools like is outpacing demonstrated returns once future revenues fail to match The Debate Humans Need
FraudGPT and DarkGPT are not and raising the specter of a finan- today’s projections. AI for Bad does not offer a final
science fiction; they are subscrip- cial bubble whose costs may ulti- verdict. Instead, it provides a frame-
tion services for cybercrime. mately fall on workers, consumers, What Needs to Change—Now work for structured, evidence-based
2. Algorithmic dehumanization— and the public. Shared accountability—Devel- debate. It is “built to be argued with”
People are reduced to data points, opers, deployers, regulators, inves- and that is precisely what we humans
with AI systems reinforcing stereo- The Benefits Are Real—But So Is the tors, workers, and users must all need. What it demands is a willing-
types and stripping away agency, Responsibility take responsibility. No more passing ness to confront both realities, quan-
often without transparency or re- The book does not call for a the buck. tified harms and real benefits, before
course. ban on AI. It acknowledges genuine Regulation that keeps pace— acting decisively. The only unaccept-
3. Psychological dependence—AI advances: productivity gains, medi- Reactive, litigation-driven responses able response is inaction.
companions and chatbots foster cal breakthroughs, accessibility im- are not enough. We need proactive, The choices we make today will
addiction, overreliance, and even provements and scientific discover- risk-tiered regulation that antici- shape whether AI becomes human-
delusional thinking, with tragic ies. But it insists that these benefits pates harm and enforces real stan- ity’s greatest ally or its unintended
cases linking chatbot interactions do not cancel out the harms. Instead, dards, especially for products reach- adversary. Let Human Goodness be
to suicide, especially among youth the authors raise the stakes: if AI is ing minors and vulnerable groups. the guiding force behind the AI, apt-
and people with mental health here to stay, humans must decide Transparency in decision-mak- ly put forward by Ravi Sharma.

www.TheIndianEYE.com