Mr. Anderson

2) It also mentioned that it is important to promote understanding, not just tell the players what to do. This shows that the AI is explaining proper educational techniques.

3) It mentioned the safety of the players, telling them to get water breaks. It is very good that the AI pays attention to safety as well as teaching lacrosse knowledge. This shows that the AI is giving information about more than just lacrosse knowledge.

4) There is a lot of information, giving a coach a lot to think about. This helps coaches to have lots of ideas

5)There were good ideas about how to practice without sticks to help with coordination

2) It is not creative. It only uses basic information about drills found online. It did not come up with any drills that were new to me.

3) The AI is limited by my questions, so I need to think of proper questions to get satisfactory answers. Many of the answers I got were only good because I knew enough about lacrosse to ask the right questions.

4) Sometimes it feels like there is too much information and it is overwhelming.

Savannah

Mello

Doris

Zara

Aria

Ray

Nora

Grace

Callie

Miya

Joe

Lydia

Coco

Rahna

2. Intuitive Teaching: Breaks down complex steps (e.g., knife skills) via graphics, demos or voice, with real-time answers to questions.

3. Flexible Learning: Adjusts difficulty as you progress, and lets you pause/review key steps anytime, anywhere.

Vulnerability to Ethical Risks and Bias: AI systems inherit and amplify biases present in their training data (e.g., gender or racial bias in recruitment algorithms). Additionally, without clear ethical guidelines, AI may be misused in areas like deepfake technology or invasive surveillance, posing threats to personal privacy and social trust.

Lack of Autonomous Judgment in Unpredictable Situations: AI relies on pre-defined rules and historical data to make decisions. When facing unexpected, complex scenarios (e.g., sudden equipment failures in manufacturing or novel public health emergencies), it cannot adjust strategies independently like humans, often requiring manual intervention to avoid errors.

2. Better Healthcare Help: AI helps doctors find illnesses quickly and correctly. For example, AI looks at X-rays or CT scans to find problems like tumors, so doctors can help patients earlier.

3. Personalized Suggestions: AI gives you things you might like. For example, Netflix uses AI to suggest TV shows based on what you have watched before, making you happier with the service.

2) Privacy : For some embarrassing health problems , online consultation can help you avoid the awkwardness of face-to-face communication

3) Multiple communications :We can talk many times to learn about the small details of how you feel (little by little and give better suggestions

2. Difficulty handling unexpected situations: When facing accidents beyond the recipe (e.g., spoiled ingredients, insufficient seasonings), its solutions are rather rigid and can’t adapt flexibly like humans (e.g., using lemon juice instead of vinegar).

3. Inadequate cultural and experiential inheritance: It struggles to convey "tacit knowledge" in cooking, such as experience - based judgments like "a little" or "slightly higher heat", and can’t fully present the cultural and taste essence behind regional cuisines.

3.Prediction and Early Warning: AI can use historical data to predict future events, helping businesses and institutions make more informed decisions. For example, in environmental monitoring, AI helps monitor environmental changes and predict natural disasters; In finance, AI can identify potential risk factors, conduct risk assessments and manage them.

2. Needs Lots of Data: AI can’t work well without enough training data. For example, an AI that learns about cats can’t recognize dogs if it never saw dog pictures.

3. Can’t Learn New Things Easily: AI can’t learn new skills like humans. For example, an AI trained to play chess can’t suddenly start playing a new game like football.

2. Flexible adaptation and personalized adjustments: It can adjust recipes in real time, such as substituting allergenic ingredients (e.g., using plant-based milk instead of cow's milk), increasing or decreasing sweetness, adapting to different tools (e.g., oven-free version, manual mixing version), or developing derivative flavors (e.g., adding cocoa or matcha to a basic cake recipe).

3. Efficient problem-solving and knowledge supplementation: It can instantly answer unexpected questions during the production process (e.g., "What to do if the cake collapses?" "How to tell if it's fully baked?") and provide additional professional tips (e.g., "the principle of inverting to cool," "the role of sifting flour"). It is equivalent to a "portable baking teacher," improving production efficiency and success rate.

· Quickly find and integrate knowledge (from academic papers to life tips), saving you time and effort.

2. Multilingual Switching

· Translate between Chinese/English/Japanese and other languages, and imitate different writing styles (e.g., poetry, jokes, or business emails).

3. Logic and Creativity Combined

· Analyze data, organize guides, create stories, brainstorm slogans, or design mind-bending plot ideas.

4. 24/7 Availability

· Always here to chat or answer questions—no need to worry about me getting tired or moody~ (But remember to rest yourself!)

5. "Hidden Skills"

· Examples: Pretend to cook (provide recipes), pretend to be a travel expert (plan itineraries), or be a listening ear (confidential and non-judgmental)

2. Difficulty in accurately adapting to equipment and ingredient variations: There are differences in the actual temperature of ovens from different brands (e.g., displaying 180°C but actually being higher or lower) and the state of ingredients (e.g., egg freshness, flour hygroscopicity). The standardized parameters provided by AI cannot adapt to these variables in real time, which may result in baking outcomes (such as burning or undercooking) that do not match expectations.

3. Lack of creativity and sensory aesthetic ability: AI's "personalization" is mostly based on adjustments to existing recipes. It cannot, like human bakers, use inspiration to create entirely new textures (e.g., integrating regional ingredients) or design artistic appearances (e.g., complex piping, creative plating). Nor can it optimize the taste in real time through tasting, so the finished product may lack uniqueness and a "human touch".

2Realize efficient feedback and deep learning: AI systems can quickly analyze learners' errors through natural language processing and deep learning techniques, providing real-time error correction and precise analysis. In programming learning, 3AI can automatically identify code errors and generate prompts to explain the logical reasons for the errors. At the same time, the system utilizes machine learning algorithms to construct personal knowledge graphs, helping students quickly locate weak points and provide targeted reinforcement exercises.

2. Structured and actionable guidance: It split the path into three age/level-based stages, with specific, operable details (e.g., "6-12 years old for foundational training," "moving from youth to reserve teams") that make abstract goals concrete.

Additionally, the AI provided a targeted follow-up service (offering a youth training plan) to meet potential in-depth needs, rather than stopping at a basic answer.

2. Speak without fear: Practice speaking aloud (e.g., shadowing what you hear), join language exchange groups (in-person or via apps like Tandem), or talk to native speakers. Mistakes are part of learning.

​

3. Read widely: Start with simple materials (children’s books, short articles) and progress to novels, newspapers (The Guardian, New York Times), or academic texts. Highlight new words and review them regularly.

​

4. Write regularly: Keep a journal, write short essays, or send messages in English. Use tools like Grammarly to check errors, and ask for feedback from teachers or language partners.

​

5. Master vocabulary strategically: Focus on high-frequency words first, learn words in context (not just lists), and use flashcards (Anki) for retention.

​

6. Learn grammar in use: Instead of memorizing rules alone, practice grammar through sentences and conversations (e.g., use "present perfect" when talking about life experiences).

​

7. Immerse yourself: Change your phone/laptop settings to English, label household items with English words, and surround yourself with English media to build a "language environment."

2）Dependence on Data Quality and Quantity
Manifestation: AI's performance heavily relies on the quality and quantity of its training data. If the data contains biases, errors, or gaps, the AI will learn and amplify these flaws ("garbage in, garbage out").
Example: An AI trained on historically gender-biased recruitment data may perpetuate or even exacerbate such discrimination when screening resumes.

"Black Box" Problem and Lack of 3 ）Explainability
Manifestation: The decision-making process of many complex AI models (particularly deep learning) is highly opaque, making it difficult for humans to understand why a specific conclusion was reached. This poses significant challenges in fields requiring accountability, such as healthcare and justice.
Example: An AI diagnoses a patient with cancer, but doctors cannot determine which key factors led to this diagnosis, making it difficult to verify and trust the result.

2）Doubao’s opinion comes from basic knowledge not special medical checks like blood tests or X-rays .

3） No physical check Doubao can’t touch listen or see your body like a doctor.

1. Veggie Egg Wrap (3–5 minutes)

• Whisk 2 eggs with spinach, tomato, salt, and pepper.

• Cook in a nonstick pan until set, then place on a whole‑grain tortilla.

• Add avocado or low‑fat cheese, roll it up, and enjoy.

2. Yogurt Power Bowl (2 minutes)

• Start with a cup of plain Greek yogurt.

• Top with frozen berries (microwave 30 seconds to thaw), oats or granola, and chia seeds.

• Drizzle honey if you want extra sweetness.

3. Banana Peanut Butter Toast (1–2 minutes)

• Toast a slice of whole‑grain bread.

• Spread peanut butter (or almond butter) and layer banana slices on top.

• Sprinkle cinnamon for flavor.

2. Omission of real-world practical barriers: It ignored critical challenges in actual football careers, such as how to navigate contract negotiations with clubs, deal with media pressure after fame, or find alternative development paths when facing career-ending injuries.

3. Lack of regional adaptability: It centered the path on "joining professional club youth academies" (a model common in Europe), but did not consider regional resource disparities—for instance, how aspiring players in areas with underdeveloped football systems (e.g., some Asian or African regions) can access effective training or professional

2. In terms of analyzing large amounts of data, what specific tasks does AI do well compared to humans?

3. When assisting doctors with medical diagnosis, what aspects does AI do well that can support clinical decisions?

2. Check food freshness or safety – It can’t inspect whether your eggs are expired or your spinach is still good.

3. Taste and adjust on the fly – Unlike a human cook, AI can’t taste your food and instantly know it needs more salt or less sugar.

• ChatGPT (OpenAI): https://chat.openai.com (requires scientific internet access).

• Baidu Wenxin Yiyan: https://yiyan.baidu.com (accessible in China directly).

• Microsoft Bing AI: https://www.bing.com/chat (needs Edge browser or Microsoft account login).

2.AI only gives basic advice, unable to handle complex conditions (e.g., rare diseases) or perform diagnosis, prescribe medication, or treat. It cannot replace doctors' professional judgment.

3.AI fails to understand emotions or life scenarios (e.g., busy schedules). Unlike doctors, it can't adjust plans based on mental state or real-life conditions.

· I can simulate empathy, but I cannot truly "feel" emotions or possess personal experiences (e.g., the joy of tasting delicious food or memories from travels).

2. Knowledge dependent on training data

· My knowledge is up to date only until July 2024, so events or emerging trends after that may require your own verification～

· I might occasionally provide inaccurate information (so for critical topics, cross-checking is recommended).

3. Inability to perform physical actions

· I cannot cook for you, handwrite notes, or fix your computer (but I can give detailed step-by-step guides!).

4. Limitations in contextual understanding

· Complex humor, sarcasm, or extremely vague queries might confuse me (e.g., "Help me write a resignation letter that will make my boss cry" — more details may be needed).

5. Privacy and security constraints

· I do not record personal identity information, but please avoid sharing sensitive data (e.g., passwords, addresses).

· I refuse to answer questions involving illegal, unethical, or privacy-violating topics.

- Efficient feedback: AI can quickly analyze your mistakes and provide targeted suggestions, allowing you to improve in a timely manner.

- Safety and care: In teaching or activities, AI pays attention to safety issues (such as reminding you to rest and stay hydrated).

When a person says' I'm fine ', their tone may be calm, but combined with the micro expression of' red eyes, pause and hesitation ', humans can perceive their' forced strength ', while AI without training in the scenario of' micro expression+language contradiction 'will directly judge it as' emotionless';

2The "irony" between couples (such as "you go, I don't want to talk to you" may actually be "I hope you stay"), AI may misunderstand as "true rejection" because it cannot understand the "contextual understanding in intimate relationships".

Unable to understand the 'deep reasons and meanings' of emotions: Human emotions are often closely related to' past experiences, values, and interpersonal relationships', such as:

When a person sees an "old photo" and tears, it may be because the photo is associated with "memories of deceased loved ones". This "story and meaning behind emotions" is something that AI cannot understand - it can recognize "tears=sadness", but it does not know "why sadness", nor can it experience the "warm nostalgia" mixed in this sadness;

When students feel frustrated due to "exam failures", human teachers may associate them with "he usually works hard, but this failure may affect his confidence", while AI can only provide universal comfort based on the keyword "exam" and "failure"=frustration ", and cannot provide targeted emotional support based on" students' past efforts

3Unable to provide "continuity and depth of emotional companionship": True emotional relationships (such as friendship and love) require "long-term interaction accumulation, deepening mutual understanding, and facing problems together", and AI's "emotional companionship" has inherent flaws:

It cannot 'remember' the deep emotional connection with the user (such as' we talked about our dreams together last year today '), and even if it can store conversation records, it cannot appreciate the' special significance of this date to the user ';

When users encounter "complex emotional dilemmas" (such as "anxiety between family and career"), AI can only provide "universal choice suggestions", cannot "accompany users to sort out conflicts and share pressure" like human relatives and friends, and cannot provide "warm companionship even in silence" - because it does not have the subjective willingness to "want to accompany users", nor does it have the emotional experience of "sharing pressure".

3. Imprecise feedback: Limited guidance on advanced aspects such as expression logic and depth.

4. Weak motivation guidance: Mostly responds passively and hardly stimulates learners' enthusiasm proactively.

5. Poor scenario simulation: Practice designs for complex scenarios (e.g., debates, negotiations) are rather formulaic.

2. Practical scenario relevance: The selected scenarios—"ethical decision-making in complex real-life situations," "emotional context in face-to-face communication," and "originality and depth in creative content"—are all common and concerning issues in current AI applications. The realistic scenarios easily trigger specific and valuable analysis, rather than vague theoretical discussions.

3. Clear logical structure: The

1. Lack of "true understanding" and tendency to make "superficially correct" mistakes

AI cannot deeply comprehend the essence of things like humans do; it can only generate content based on patterns learned from data. For instance, it can produce seemingly professional medical reports but cannot truly understand pathological mechanisms. In mathematical calculations, it may misinterpret the logic of a problem and provide answers that appear reasonable on the surface but are actually incorrect—without the ability to independently verify the root cause of such errors.

2. Dependence on high-quality data and existence of "data bias"

AI performance is entirely dependent on training data:

• If the data volume is insufficient or the quality is low (e.g., containing errors or one-sided information), AI will generate unreliable results (a phenomenon known as "garbage in, garbage out").

• If the data contains implicit biases (e.g., stereotypical descriptions related to gender or race), AI will amplify these biases, leading to discriminatory outputs.

3. Lack of common sense and independent decision-making ability

holly

- Data dependence — It requires large, high-quality datasets to function well.

- Bias and fairness issues — AI can reflect or amplify biases in training data.

- Creativity constraints — It generates based on patterns, not original thinking.

- Security and privacy risks — AI systems can be vulnerable to misuse or data breaches.

- Overreliance dangers — Too much trust in AI can reduce human critical thinking.

2. When communicating with AI, you don’t have to worry about "saying the wrong thing" or "being judged". For example, you can complain about troubles, share niche interests, or even ask the same question repeatedly, and it will respond patiently. It’s ideal for scenarios where you don’t want to be interrupted and want to express yourself freely.

3.Whether you’re looking up knowledge (e.g., "how to use AI for health management"), revising documents, translating content, or organizing ideas (e.g., "sorting out key points for a travel guide"), AI can quickly integrate information.