SUMMATIVE ASSESSMENT FOR THE UNIT «STEM»
Subject: English
Grade: 11
Total marks: 12
Learning Objectives
- 11.2.4 Understand implied meaning in unsupported extended talk on a wide range of general and curricular topics, including talk on a growing range of unfamiliar topics.
- 11.4.4 Read a wide range of extended fiction and non-fiction texts on a variety of more complex and abstract general and curricular topics.
- 11.5.1 Plan, write, edit and proofread work at text level independently on a wide range of general and curricular topics.
- 11.3.2 Ask and respond with appropriate syntax and vocabulary to open-ended higher-order thinking questions on a range of general and curricular topics, including some unfamiliar topics.
Task 1. Reading (LO 11.4.4)
Instructions: Read the text carefully. Decide if the statements (1-4) are True (T) or False (F). Choose the correct answer for each statement.
The Ethical Dilemma of Artificial Intelligence in STEM
The integration of Artificial Intelligence (AI) into Science, Technology, Engineering, and Mathematics (STEM) fields is accelerating at an unprecedented pace. From drug discovery algorithms that can predict molecular interactions to autonomous systems managing complex power grids, AI's potential seems limitless. However, this rapid advancement brings forth profound ethical questions that the STEM community must address.
One primary concern is bias in algorithmic decision-making. AI systems learn from vast datasets, which often contain historical and societal biases. For instance, a facial recognition system trained primarily on one demographic may perform poorly on others, leading to unfair outcomes. Similarly, AI used in hiring for tech companies might inadvertently perpetuate gender or racial disparities if the training data reflects past inequalities. The implication is clear: the objectivity of STEM is challenged by the subjective data we feed into our machines.
Another significant issue is accountability. When an AI-driven medical diagnosis system makes an error, who is responsible? The software developer, the hospital administering it, or the algorithm itself? Traditional STEM disciplines have clear chains of responsibility, but AI, especially with self-learning capabilities, blurs these lines. This creates a "responsibility gap" that lawmakers and ethicists are struggling to fill.
Furthermore, the environmental cost of AI is often overlooked. Training sophisticated AI models requires immense computational power, leading to a substantial carbon footprint. A single large model's training can emit as much carbon dioxide as five cars over their entire lifetimes. Thus, the pursuit of technological advancement in STEM must be balanced with sustainable practices.
Addressing these dilemmas requires a multidisciplinary approach. It is no longer sufficient for STEM professionals to be experts only in their technical fields. They must engage with ethics, sociology, and law to develop AI that is not only intelligent but also fair, accountable, and sustainable. The future of STEM depends not just on what we can build, but on what we should build.
Statements:
- The text suggests that the main problem with AI in STEM is its high financial cost. T / F
- According to the text, AI can sometimes worsen existing social inequalities. T / F
- The "responsibility gap" refers to the lack of powerful computers to train AI models. T / F
- The author implies that future STEM education should include non-technical subjects like ethics. T / F
Assessment Criteria for Task 1
Marks: 4 (1 mark for each correct answer).