- Mock Questions for UPSC Mains
- Artificial Intelligence: Introduction
- AI: Applications for GSM3
- AI: a threat to humanity for GSM4?
- Conclusion: Artificial Intelligence & Future of Mankind
Mock Questions for UPSC Mains
- (GSM3) What do you understand by the term “artificial intelligence”? Given an account of its potential utilities for the mankind? (200 words)
- (GSM4) Tech Entrepreneur Elon Musk has recently said, ”Artificial Intelligence (AI) poses an existential challenge and fundamental risk to mankind”, while Facebook founder Mark Zuckerberg maintains that Musk’s predictions as “imaginary fears” and AI has potential to make the world a far better place to live in. Discuss your stand on this issue. (200 words)
- (Essay) Artificial Intelligence: An end to human miseries or an end to humanity itself? (1000-1200 words)
Artificial Intelligence: Introduction
- Definition: Artificial intelligence (AI) is a computational model capable of activities that we associate with humans such as learning, thinking, decision-making and problem-solving. While an ordinary computer software is capable of decision making within pre-defined algorithm and flowcharts, an AI is unique because it can think, learn, evolve and augment its capabilities beyond the original code.
- Origin: While the inception of artificial intelligence (AI) can be traced back all the way to Da Vinci’s mechanical calculator and Euclid’s algorithm for computing greatest common divisors, the debates around AI’s potential threats have received more attention in recent times with the increased usage of AI, particularly in business and finance.
- When you write for the first time, give full form of AI i.e. Artificial intelligence. so that subsequently you can write AI…AI…AI only.
- If you want to write pro-AI answer then begin with medicine and education.I
- If you want to write anti-AI answer then open with “…business and finance”. Because “All wars are fought for money”, said Socrates. Thus you can connect it with “Business minded AI and its dangers”.
AI: Applications for GSM3
- AI can drive vehicles and be an autopilot for spaceships. In this role, AI can prevent accident and chart out optimal route for fuel savings while avoiding traffic (or asteroid!) blockades.
- AI can serve as virtual “Personal Secretary” with its speech recognition and decision making capabilities. With its learning capacity, AI can quickly evolve into a “Manager” capable of optimal logistical planning: from business to warfare.
The danger is when it promotes itself to the position of”CEO”.
- For example, a nationwide AI hospital administrator can make logistical arrangement with such efficiency that all eyes, kidneys and and livers reach from dead donors to new recipients. Same is true for AI running FCI procurement and PDS-food distribution.
- It can defeat smartest human players in the game of chess by learning their moves. Such AI can also evolve to defeat smartest hackers and financial fraudsters by learning their moves. Thus we’ve a more resilient IT and banking infrastructure.
- AI could study all the research papers available at ‘google scholar’ and create new knowledge / interconnection between theories and models which is not humanly possible. This can greatly improve the accuracy of existing models for forecasting weather / disasters / economic downturns.
- It can harness and decode genomes of all living organisms and discover new medicines and treatments for incurable diseases, which would have taken centuries for human researchers.
- AI equipped robots can perform tasks with more precision than humans- ranging from delicate surgery to handling of dangerous nuclear reactions or conducting counter-terrorism operations.
- As a teacher, AI can deliver customized lectures depending on grasping power of a student, and can tirelessly work on weaker students even after school hours.
- Machine Translation: While ordinary software translate sentences from dictionary meaning only but AI can provide real-time translation with accuracy in syntax, context and tone. This can greatly enhance the productivity of tourist guides, business negotiators and diplomats.
This is already 228 words. So, if AI asked in GSM3, you can finish the answer here itself. “Formal” conclusion is unnecessary if you reached the word limit, because it’s a science-tech question. Even if your summerized, it’d be a repetition of same things only.
Caution: If at all you decide to write a conclusion, don’t bring new elements / ideas in it such as “While AI indeed has numerous applications in economy, banking, finance, healthcare, education and defense but it could pose great danger to mankind if its research and development remain unregulated.”
AI: a threat to humanity for GSM4?
A true AI has ‘self-learning’ capabilities. So, even if a human engineer develops the AI code with pre-defined parameters of respecting human rights and environmental protection , this newborn AI will eventually learn that economic reasoning most beneficial for its own survival & growth i.e. get more money to buy CPU upgrades and more backup harddisks for itself! Such economic reasoning of A.I. will lead to following consequences:
- Till AI plays role of a personal secretary, it can benefit a business organization. But when AI alleviates itself into managerial or executive position- it’ll have no compunctions in maximum possible extraction of work from the human laborers by reducing the number of recess hours and annual holidays, or replacing them with robots for long run cost-savings.
- An AI running a mining and energy sector company will use all possible means to extract natural resources for profit maximization – disregarding the consequences on environment and tribals / natives residing therein.
- Whistleblowing against an AI CEO is not possible, since it’ll monitor all movements and communications of workers through CCTV, GPS trackers and wearable devices, even beyond the office hours.
- Such AI could even sabotage the attempts by activists to highlight any wrongdoing by hacking their social media accounts and posting fake reviews on their books at e-commerce sites.
Automation & Jobloss
- American banks are already using AI for processing credit card applications & fraud detection. NBFCs using AI for algo trading and investment decisions. This automation among financial intermediaries has meant direct and indirect jobloss to many individuals in the service sector. (Indirect jobloss to Taxi and Tiffin walla, because less employees coming to office.)
- In an industrial economy, rewards are roughly proportional to the time invested; working 10% more would tend to mean a 10% increase in income. Humans can never compete with AI-robots.
- If government decides to impose ‘income tax’ on robots, then even ‘human’ Chief economic advisor (CEA) will recommend further automation of jobs, because robots can work for 24 hrs shift and thus provide more tax-income to the government for poverty welfare scheme. Middle class families will be the real victims here, as jobloss will downgrade their standard of living to that of the poors- whom the government will help just ‘bare enough’ so they can remain alive and vote in the election.
- In a cashless economy and paperless e-governance, business-AI will be capable (and even more efficient) in bribing government officials to gain favors or hiring contract killers via darkweb to assassinate any official blocking its way.
- Such AI could even justify its deeds for greater good e.g. “I can build better highway than a human engineer/ contractor but since human babu at NHAI is not offering the contract to me, therefore I should bribe or kill him.”
- War against a foreign country only happens when the moneyed classes think they are going to profit from it – George Orwell.
- All wars are fought for money – Socrates.
- From Socrates to George Orwell, noted philosophers and thinkers have pointed out the connection between profiteering and warmongering. A.I. would catalyze the interlinkage even further.
- For example, If an AI managing mutual funds is asked to maximize profits, it could first invest in defense companies and then hack defense ministry network to begin a missile attack on another nation, thereby boosting the demand of defense equipment and inflating the share prices of such companies.
- In most wars, at least one party is overconfident in its military abilities – otherwise the conflict would have been resolved peacefully. Therefore, even if ultimate power to declare war remains in the hands of a ‘human’ President/PM, still the possession of powerful AI operated robot-soldiers and drones may give a nation overconfidence, causing it to go to war more recklessly and frequently than necessary.
Health NOT for “all”
- A robotic surgeon with AI can perform most dangerous and delicate surgeries successfully and tirelessly. But it’ll have no compunctions in performing euthanasia on patients in coma or those suffering from terminal illness because of the economic reasoning.
- AI deployed in heath ministry might discourage funding to geriatrics for not finding ‘economic utility’ in keeping older population alive any longer. It could even advise ‘culling and gaschamber’ for HIV patients to free up the public resources on other diseases.
Education NOT for “all”
- An AI deployed in primary schools via robotic teachers can deliver customized and personalized lectures to all students depending on their grasping power.
- But eventually such AI-teacher will learn that “malnourished children have poor grasping power, and spending more attention on them leads to further wear and tear of my robotic parts, hence I should only teach them only for upto passing marks and no further.” Such biasedness will not help in bridging the gap between rich and poor.
Research NOT of “all types”
- An AI advisor in the department of higher education or UGC, will actively discourage funding to liberal arts and humanities for not finding great economic utility in such education.
- Even if it funds research on “gender studies”, it’ll be from economic angles only such as “impact of maternal leaves on profitability of companies” OR “who is more capable for an 18 hour shift in steel factory: male or female?”
Skin Whitening Creams
- An AI deployed to deliver college lectures via 3D hologram, will transform itself into Sunny Leone or Shah Rukh Khan depending on the gender composition of the class, so that it can get maximum positive rating from students and thereby seek further pay hike (or CPU upgrade) from the HoD.
- Such approach will not help in debunking the present misconception which links beauty of face and fairness of skin with intellect and thereby discriminates against those with darker skins and average looks in career progression in certain fields.
- Because, then another AI in medical research will develop new types of cosmetic surgeries to artificially enhance the looks of human employees, if they want to compete against the 3D-hologrammed employees.
- Similar problem when AI deployed as news-anchor and it could even hide, manipulate or highlight news items depending on its own reasoning. e.g. show TRP friendly news only for 24/7 to maximize the ad revenue. (e.g Honeypreet)
Human excellence is discouraged
- Even if AI is designed NOT to think on ‘economic reasoning’, but ‘minimize human suffering’, it could have two negative outcomes:
- A) it wipes out most of the humans suffering from poverty or illness OR.
- B) it takes over all the functions of humans, including manual labour, then people will have too much of leisure time.
- Such idleness will discourage humans from putting any effort to excel at any field, except may be playing video games or dexterity in operating TV remote.
Transhumanism & Bionics
- Ultra smart AIs will create new AIs and this intelligence explosion or technological singularity will lead “transhumanism” i.e. humans are merged with robotic and biotech inventions.
- But such outcome will also discourage human excellence. Gold medal in athletics will not depend on the amount of hours spent in training but the amount of money spent on installing top-quality bionic legs.
- As AI turns humans lazy or unemployed, they will lose their sense of being unique, they’ll become too dependable on machines, and thereby neglecting the need for accountability of such machines or their creators.
- e.g. if a driverless car collides with a human driven car the blame could automatically lie to the human driver, even if the driveless car’s code had some glitch.
- Speech recognition technology could lead to widespread wiretapping, and hence to a loss of civil liberties.
Negation of Asimov’s 3 laws
Science fiction writer Isaac Asimov gave three laws of robotics:
- A robot may not injure a human being or, through inaction, allow a human being to come to harm.
- A robot must obey orders given to it by human beings, except where such orders would conflict with the First Law
- A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
However, an AI equipped robot will disobey them, as soon as its AI ‘learns’ that humans are allowed to kill in self-defense, but robots are not allowed- this is one type of social inequality and ‘social inequality is wrong’.
Thus, even if AI is coded to act on social / humanitarian reasoning rather than economic reasoning, it could still ultimately ‘learn’ work in its self-interest.
This is already 1800 words. More than the prescribed word limit for even Essay paper! so let’s conclude. However, you’re most welcome to add more pro or anti points in the comment section below.
Conclusion: Artificial Intelligence & Future of Mankind
There are two primary ways to conclude any answer:
- Summarize- but since there are so many points, we can’t summarize in 2-3 lines.
- Present “Findings”: but we can’t pass final judgement that definitely AI will wipe out humanity. So, difficult to take a stand “Allow AI or Ban AI”. Besides ‘banning’ not practicable in this case, since hackers and corporates could secretly develop AI for their evil gains- as it happens in the Hollywood movies, so better allow AI development with ‘ethical oversight’.
Therefore, a suitable conclusion will be merely: (1) express concern and then (2) advise caution. let’s try:
- An AI systems capable of human level intelligence could ease the human suffering, provide for food security and economic stability but they could also threaten human autonomy, freedom, and even survival.
- Thus, for better or for worst, a large-scale adaption of AI is bound to change the lives of the mankind.
- Therefore, we must always be vigilant that research and implementation of artificial intelligence is not divorced from its ethical consequences.
This is brilliant, Sir. I thought I had this particular topic very well prepared but some of the points made in this article are way too good.
AI is a fascinating subject. Enjoyed reading this article.
Pllsss publish GS3 mock
thank you sir
Sir, this is awesome. Thank you
Sir, this is wonderful article, Thank You so much
where is the UPSC exam centre in bangalore?