UPSC Prelims Practice Questions — AI doomsaying

Q1. With reference to the Independent International Scientific Panel on Artificial Intelligence, consider the following statements: Which of the statements given above is/are NOT correct?

  1. It was established by a United Nations General Assembly resolution adopted by consensus on 26 August 2025, building on the Global Digital Compact.
  2. Its mandate extends to evidence-based scientific assessment of artificial intelligence in both the military and non-military domains.
  3. It is required to produce an annual report, and may additionally issue thematic briefs and provide updates to the General Assembly up to twice a year.
  4. It is empowered to lay down binding technical standards for artificial intelligence systems and to enforce compliance by Member States.
  • A. 1 and 3
  • B. 2 and 4
  • C. 1, 2 and 4
  • D. 3 only

Q2. Secretariat support to the Independent International Scientific Panel on Artificial Intelligence is coordinated by which one of the following United Nations entities?

  • A. The International Telecommunication Union, the UN specialised agency for information and communication technologies
  • B. The UN Educational, Scientific and Cultural Organization, custodian of the Recommendation on the Ethics of AI
  • C. The UN Office for Digital and Emerging Technologies, headed by an Under-Secretary-General
  • D. The UN Institute for Disarmament Research, the UN's autonomous research body on security technologies

Q3. Consider the following statements comparing the Independent International Scientific Panel on AI with the Secretary-General's High-level Advisory Body on Artificial Intelligence: Which of the statements given above is/are correct?

  1. The Scientific Panel has 40 members, whereas the High-level Advisory Body comprised 39 members drawn from 33 countries.
  2. The creation of an international scientific panel on AI and of a global dialogue on AI governance was itself among the recommendations of the High-level Advisory Body's final report, 'Governing AI for Humanity'.
  3. Unlike members of the High-level Advisory Body, who served in their personal capacity, members of the Scientific Panel sit as representatives of their national governments.
  • A. 1 only
  • B. 1 and 2 only
  • C. 2 and 3 only
  • D. 1, 2 and 3

Q4. Which one of the following correctly describes the leadership arrangement of the Independent International Scientific Panel on Artificial Intelligence?

  • A. It is headed by a single Chair, Yoshua Bengio of Canada, who holds exclusive authority over the content of all Panel reports
  • B. It is chaired ex officio by the Secretary-General's Envoy on Technology, with all forty members drawn entirely from academic institutions
  • C. It is chaired on an annual rotating basis by one expert nominated by each of the five United Nations regional groups in turn
  • D. It has two Co-Chairs, Yoshua Bengio of Canada and Maria Ressa of the Philippines, with members serving in their personal capacity

Q5. Which one of the following statements correctly describes the institutional servicing of the Global Dialogue on Artificial Intelligence Governance?

  • A. It is serviced exclusively by the UN Educational, Scientific and Cultural Organization, which alone determines the agenda of every annual session
  • B. It is serviced by a joint secretariat comprising ITU, UNESCO, the UN Office for Digital and Emerging Technologies and the Executive Office of the Secretary-General
  • C. It is serviced exclusively by the International Telecommunication Union, which alone convenes all Member States for each annual session
  • D. It is serviced by an autonomous treaty secretariat funded entirely by assessed contributions from participating Member States

Q6. Consider the following statements about how AI existential-risk claims made by industry figures in 2023 compare with those made in 2026: Which of the statements given above is/are correct?

  1. The open letter signed by hundreds of employees of leading AI companies in July 2026 called for an acceleration of frontier model training runs.
  2. In 2023 the chief executives of OpenAI, Anthropic and Google DeepMind endorsed the position that extinction risk from AI should rank alongside risks such as pandemics and nuclear war.
  3. The chief executive of Anthropic has publicly put the probability of things going 'really, really badly' with artificial intelligence at between 10 and 25 per cent.
  • A. 1 only
  • B. 1 and 2 only
  • C. 2 and 3 only
  • D. 1, 2 and 3

Q7. With reference to persons associated with safety-related departures from leading artificial intelligence laboratories, consider the following pairings: Which of the above is/are NOT correctly matched?

  1. Jan Leike — resigned from OpenAI in 2024 over safety concerns
  2. Daniel Kokotajlo — departed from OpenAI in 2024 over safety concerns
  3. Jacob Coxon — quit Anthropic in September 2026, saying the firm and its competitors were 'gambling with our lives'
  4. Evan Hubinger — resigned in 2026 as Anthropic's head of safety, warning that 'the world is in peril'
  • A. 1 and 2
  • B. 2 and 3
  • C. 1, 3 and 4
  • D. 4 only