UPSC Prelims 2020 Q38: Science & Technology – Computer & Information Technology
- (a) 1, 2, 3 and 5 only
- (b) 1, 3 and 4 only
- (c) 2, 4 and 5 only
- (d) 1, 2, 3, 4 and 5
Answer: (b) 1, 3 and 4 only
Explanation
Artificial Intelligence Capabilities: Current State (2020)
| Application | AI Can Effectively Do? | Status 2020 |
|---|---|---|
| 1. Reduce electricity consumption | YES (optimization algorithms) | YES |
| 2. Create meaningful stories + songs | PARTIAL (can generate, quality debated) | LIMITED |
| 3. Disease diagnosis | YES (image recognition, prediction) | YES |
| 4. Text-to-Speech conversion | YES (well-established) | YES |
| 5. Wireless energy transmission | NO (physics, not AI) | NO |
AI capabilities in 2020:
- YES Energy optimization (machine learning)
- ? Creative content (GANs/transformers - quality debated)
- YES Medical diagnosis (deep learning, image recognition)
- YES TTS (neural networks)
- NO Wireless energy (not an AI problem - physics)
Answer: (b) 1, 3 and 4 only - Excludes creative content (2) which is contestable, and wireless energy (5) which is physics, not AI.
Answer: (b) 1, 3 and 4 only.
Why each option is right or wrong
- (a) Incorrect. Wireless energy transmission is not an AI capability.
- (b) CORRECT. AI can effectively do: (1) energy optimization, (3) disease diagnosis, (4) text-to-speech. Statement 2 (creative content) is debatable; Statement 5 (wireless energy) is physics, not AI.
- (c) Incorrect. Wireless energy is not AI; Energy optimization is AI-solvable.
- (d) Incorrect. Wireless energy transmission is not within AI domain; it's electromagnetic physics.
Key point
AI 2020 capabilities: Optimization (energy), Diagnosis (medical), TTS (speech synthesis) = proven. Creative content = emerging but quality debated. Wireless energy transmission = physics problem, not AI.
Remember
With the present state of development in 2020, Artificial Intelligence can effectively do: (1) Bring down electricity consumption in industrial units (optimization), (3) Disease diagnosis (image recognition), and (4) Text-to-Speech Conversion. Statements 2 (creative content quality is debatable) and 5 (wireless energy is physics) are not AI capabilities.
Common mistakes
- Assuming AI can solve any problem labeled 'technology.' Wireless energy is electromagnetic physics, not algorithmic.
- Overestimating creative AI. While GPT-2/3 can generate text, quality for 'meaningful stories' is subjective and mixed.
- Missing that medical diagnosis (radiology, pathology) is one of AI's clearest successes as of 2020.
Exam tip
AI vs physics: AI = optimization, pattern recognition, classification. Physics problems (energy transmission, materials) require domain-specific research, not machine learning. Conflating them is category error.
Sources
- Yoshua Bengio, Ian Goodfellow, Aaron Courville - Deep Learning (MIT Press)
- Stuart Russell and Peter Norvig — Artificial Intelligence: A Modern Approach (Pearson)