AI and Wikis - NGUYEN DUC DUY

Assignment 1

I generally prefer to use AI, such as ChatGPT, because it is fast, convenient, and easy to understand. AI can quickly summarize complex topics and present them in a simple way, which is helpful when I need a quick answer or when learning something new. For example, when studying English grammar, I can ask AI to explain a rule and give examples instantly instead of reading long explanations. This makes my learning more efficient and less overwhelming.

However, my choice depends on the situation. While AI is good for speed and simplicity, Wikipedia is more reliable for academic work. It provides detailed information, clear structure, and references that can be checked. For example, when writing assignments or checking important facts, I prefer Wikipedia to ensure accuracy.

Overall, I usually start with AI to understand the topic quickly, then use Wikipedia for more reliable information.

Assignment 2

I think AI should be used for Wikipedia, but in a limited and careful way.

The Slate article shows that AI can quickly generate a full article, which is useful for saving time. However, it can also create fake information and references that look real but do not exist. This is dangerous because people may trust incorrect information.

The Vice article presents real experiences from Wikipedia editors. Some users post AI-generated content, but other editors must spend a lot of time checking and fixing it. Sometimes it is even faster to write a new article than to correct AI mistakes. This can also reduce trust between editors.

In my opinion, AI should only be used as a helper. It can summarize information or suggest ideas, but humans must always verify facts and sources. Using AI without checking is risky, especially in academic work.

Overall, AI should support human editors, not replace them.

Bonus

When I asked AI this question, it gave a balanced answer, mentioning both benefits and the need for human checking. However, the Slate and Vice articles focus more on real problems, such as fake citations and extra work for editors.

This shows that AI tends to give general and positive answers, while real sources highlight practical risks.

Assignment 3

Wikipedia deals with AI in a strict and organized way.

First, AI has been used for a long time through bots and tools like ClueBot NG, mainly for support tasks. However, modern AI can generate content that looks correct but may contain errors.

Second, Wikipedia has clear rules: AI cannot freely create or publish articles. It can only assist with tasks like grammar or translation, and humans must review everything.

Third, there is a WikiProject AI Cleanup that focuses on finding and fixing AI-generated content. This shows that AI content is already a real issue.

What I find most interesting is how cautious Wikipedia is. It does not reject AI, but it prioritizes accuracy and human responsibility. In some cases, AI even creates more work, which shows that reliability is more important than speed.

Assignment 4

AI is very good at structure. It can quickly create a Wikipedia-style article with sections like “History” and “Impact.”

However, problems appear in the content. For example, AI often gives general statements without clear or reliable citations, while Wikipedia includes detailed references that can be checked. AI also lacks precision, such as giving vague timelines instead of specific facts.

There is also a difference in depth. AI tends to summarize information, while Wikipedia provides more detailed explanations, statistics, and multiple viewpoints.

Overall, AI is useful for creating a first draft, but it is not reliable enough for a final article.

Assignment 5

STORM uses a more complex process than ChatGPT. It simulates multiple “AI editors” with different perspectives, collects information, and builds an outline before writing. This results in more detailed and balanced articles.

In contrast, ChatGPT generates content immediately. It is faster but often more general and less detailed.

However, both systems have similar problems, such as hallucinated citations. Even STORM can create sources that look real but are not.

STORM is better in process and depth, but not necessarily more trustworthy. Both still require human verification.

Assignment 6

AI can help in several practical ways, such as improving writing, summarizing sources, and translating content.

For example, it can turn simple sentences into a more formal and neutral style suitable for Wikipedia. It can also summarize long sources, helping users understand key ideas quickly.

However, some uses are less practical. For example, generating a full article is risky because AI may include incorrect information or unreliable sources. Suggested references may also need to be checked manually.

Overall, AI is most useful as a support tool, not as a main author.

Assignment 7

Grokipedia is better in terms of speed and readability. It provides smooth and easy-to-understand explanations in a continuous format.

However, it often lacks strong sources and precise details. For example, it may give general statements without citations, while Wikipedia provides specific data and verifiable references.

Wikipedia is also more neutral and balanced, especially on controversial topics, because it is reviewed by many editors.

Overall, Grokipedia is useful for quick understanding, but Wikipedia is more reliable for academic purposes.