Major Milestones in AI Evolution AIE Practice Questions (Page 7)

Questions 31–35

1. 31application · easy

   A data science team is preparing a presentation on the history of AI for new interns. They want to show a timeline that correctly sequences the following milestones: (1) the development of deep learning models like AlexNet, (2) the creation of expert systems like MYCIN, (3) the proposal of the Turing test, and (4) the rise of generative models like GPT. Which chronological order should the team present?

   • AGenerative models, deep learning, expert systems, Turing test.
   • BDeep learning, Turing test, expert systems, generative models.
   • CTuring test, expert systems, deep learning, generative models.
   • DExpert systems, Turing test, generative models, deep learning.
   Reveal Solution Discussion Select an answer first
2. 32expert · hard

   A research institute is planning a long-term AI strategy. They want to avoid repeating the mistakes of past AI winters. Which approach is most aligned with lessons learned from those periods?

   • ABalance ambitious research with practical applications that deliver incremental value.
   • BFocus exclusively on high-risk, high-reward projects to achieve breakthroughs quickly.
   • CAvoid any research that does not guarantee immediate commercial success.
   • DRely solely on government funding to ensure stability.
   Reveal Solution Discussion Select an answer first
4. 33expert · hard

   A government agency is planning to modernize its legacy document-processing system. The current system uses rule-based expert systems that have been in place since the 1990s. The agency is considering replacing it with a deep learning model to handle a wider variety of document formats. However, the agency must maintain a complete audit trail for every decision due to legal requirements. The project lead is concerned about the 'black box' nature of deep learning. Which approach best balances the need for accuracy on diverse formats with the legal requirement for explainability?

   • AKeep the expert system for all decisions, as it is the only approach that can provide a complete audit trail.
   • BUse a deep learning model and train it to output a confidence score, and consider that score as the explanation.
   • CUse a hybrid approach: deploy the deep learning model for initial classification, but route any decision with low confidence to the expert system for final review and logging.
   • DReplace the expert system entirely with a deep learning model and rely on post-hoc explanation tools to generate audit logs.
   Reveal Solution Discussion Select an answer first
5. 34application · medium

   A tech journalist is writing an article about the 'AI winter' periods. Which statement accurately describes the impact of these periods on AI development?

   • AAI winters permanently halted all AI research, and no progress was made until the 2012 deep learning breakthrough.
   • BAI winters occurred only in the 1970s and had no effect on later AI development.
   • CAI winters were periods of reduced funding and interest, but research continued in areas like expert systems and neural networks.
   • DAI winters were caused by the success of expert systems, which made AI too expensive for commercial use.
   Reveal Solution Discussion Select an answer first
6. 35expert · hard

   A company is evaluating whether to adopt a generative AI system for content creation. The system is based on a large language model that was trained on a vast corpus of text. The company's legal team is concerned about the potential for the model to generate copyrighted material. The team lead notes that the model's training data includes copyrighted text, but the model does not reproduce it verbatim. Which historical milestone is most relevant to this concern?

   • AThe development of expert systems, because they encode human knowledge and can be audited.
   • BThe rise of deep learning and generative models, because they learn patterns from large datasets and can produce outputs that may inadvertently resemble copyrighted material.
   • CThe 'AI winter', because it shows that AI systems can fail when expectations are too high.
   • DThe Turing test, because it evaluates whether a machine can imitate human behavior.
   Reveal Solution Discussion Select an answer first