Post 7

AI image generation is honestly wild. It’s fast, it’s fun, and once you start playing with it, it’s kind of hard to stop. But these models are all being trained on billions of images scraped from the internet such as your photos, artists’ portfolios, news footage, and usually without anyone’s permission. It’s not about that AI can copy a style; it’s that it can swallow someone’s face, their work, and their memory without asking, then spit it back out as something “new.” That’s not innovation, that’s extraction.


And it gets way worse when people use this tech to generate images of someone who’s dead, such as like Kobe Bryant——honestly it’s kind of shame because it was a trend in China. I totally understand that people miss him, but Kobe can’t consent to having his likeness dragged out of old footage and turned into a digital being because he is dead. It’s super disrespectful.


So how can we actually fix this? I think there should be a regulation for data that used to train AI, as we discussed a lot. First, these models need to be trained on opt-in data, not just everything scraped off the web. Second, every AI image needs a permanent watermark or label that can’t be cropped out, and platforms need to actually enforce it. And third, there should be a hard rule: you don’t generate identifiable images of real dead people without explicit family consent. So we need to stop treating “we can do it” as the same thing as “we should do it.”.

Post 6

AI is super, super useful. It is probably beyond the level of the invention of Google. It’s fast, easy to use, even easy to read——it summarizes everything for you. It’s like addiction. Once you start using it, it’s hard for you to stop. It’s not a bad thing to do, I personally use AI every day and even for my own study process (ethically). But at the same time, problems occurs. The single most important generative AI issue the public needs to understand is that AIs can so called “invent” facts, citations, people, or events and send you them with real information. The issue is not AI makes mistakes, but it makes mistakes confidently, often combined with actual sources, makes it difficult to distinct.

There was once when my mom brought me information that were generated by AI, which says that she doesn’t need a passport to get on a plane but with abcde they will still allow you to get on the plane. Absolutely not possible, and she even encouraged me to try it out May 13th when I get back home.

we really need AI in schools and workplaces to be trained to produce outputs such as drafts requiring verification, not answers requiring trust. We have to pay more attention because hallucinations is not avoidable. Also, we need to realize that there is a clear line between “sounds right” and “is right”.

Post 5 Academic AI

AI for academic using was never an issue before people realize that students can use AI to write great essays. I still remember the last semester of my high school when my friend wrote an entire essay with AI without being revised by himself, and got an A for his final grade.

In short, the concept of too much means when you are not ready to handle the knowledge yet, or too lazy to write the entire or both, which is the most common use of AI for students now. Once AI starts replacing our cognitive work that builds expertise, it is too much. Dinsmore & Fryer argue that human learning depends on three forces, which are Knowledge, Strategies and Interest.AI might harm our process and willingness of learning if we are skipping any of these three processes. Because it is a linear progress, acclimation to competence to expertise. If we use AI inappropriately, AI will become a shortcut that destroy the entire mental process of integrating your knowledge.

Source: Dinsmore, Dan L., and Luke K. Fryer. “What Does Current GenAI Actually Mean for Student Learning?” Learning and Individual Differences, vol. 125, Jan. 2026, p. 102834, https://doi.org/10.1016/j.lindif.2025.102834.

Post 4

Honestly, I would not say that any AIs do have any sort of creativity. I think creativity does not just about Meaningfulness and novelty, but also the willingness to express. I think one example from the activity on Tuesday shows this. I used this prompt: “Good morning Chat, Please write a poem in the style of Allen Poe, including his writing skills such as figure of speech, metaphor, symbolism, and narration. Take the Raven by Allen Poe as an example.”

And Chatgpt gave me this:

The Lantern and the Tide

Upon a midnight waning, when the stars were dimly straying,
And the wind spoke low and hollow through the ribs of rotting pine,
I sat in silence keeping, though I knew no gentle sleeping
Would descend upon a spirit so entangled yet as mine—
For a thought, long dead yet breathing, pressed its weight upon my spine,
Whispering, “You are mine.”

From the hearth no flame was turning, only ash in quiet yearning
Gave a dull and ashen glimmer, like an eye that would not close;
And the walls, with shadows bending, seemed to breathe a slow pretending,
As if memory itself had learned the art of silent throes—
As if time, with careful fingers, stitched its grief in hidden rows,
Where no daylight ever goes.

Then—there came a sound uncertain, like a hand upon a curtain,
Softly drawn yet filled with meaning none could ever dare define;
Not a knock of mortal making, nor a timber’s weary breaking,
But a pulse—alive, insistent—beating somewhere out of line.
“Who now walks,” I whispered faintly, “through the ruins of my mind?”
Came the answer: “Only thine.”

To the door I turned, unsteady; all the air grew thick and heavy,
And the lantern by my table flickered thrice, then guttered low;
When I opened—there was nothing but the dark, complete and muffling,
Save a tide of distant murmurs from a sea I did not know.
Yet within that void unbroken, something stirred, and seemed to grow—
Not of flesh, nor shade, nor woe.

Back I stepped, my breath retreating, heart and hollow rhythm meeting,
Till the chamber shrank to nothing but a stage for dread alone;
And upon the rafters leaning, where the blackest dark lay screening,
Sat a shape no light could capture, carved in silence, cold as stone.
Eyes like wells without a bottom fixed themselves upon my own—
And it spoke in monotone:

“Do you seek what once was taken? Do you walk though you are shaken?
Do you bind yourself to echoes that no dawn shall ever free?”
Thus it uttered, never lifting, never faltering nor drifting,
Like a verdict etched in iron, cast beyond eternity.
And I, though weak with knowing, asked the thing that haunted me:
“Shall there yet be mercy?”

Then the lantern died completely, and the dark grew far too neatly
Round the corners of my reason, folding thought in endless night;
Still the voice, unspent, unbroken, left no comfort in its token—
Only this, in endless echo, stripped of hope and stripped of light:
Not a cry, nor curse, nor pleading—only cold, unyielding rite:

“Evermore—despite.”

And the tide, though never nearing, I could hear it—ever searing,
Washing names from every surface, wearing time itself away;
While the figure, ever keeping watch above my spirit’s weeping,
Cast no shadow, claimed no substance, yet refused to fade or stray.
Thus I sit, with thought unending, where the night outlives the day—
And the tide will have its way.

You can see that the poem itself is great, but when we dig into the actual meaning of it: barely anything. AIs do not have any actual ideas and things to produce but to make something that seems to be right. The article from Freethink say that “AI doesn’t feel tension. It doesn’t labor through ambiguity. It doesn’t explore.” which I think is absolutely correct. AI does not create, but to gather old information together and form a “seems like a new” one.

This activity kind of motivates me to think more and write more with my own hands (and keyboard), but AI is still a great tool to develop ideas when stuck.

Bhatia, Ashish. “The Artifact Isn’t the Art: Rethinking Creativity in the Age of AI.” Freethink, April 4, 2025. https://www.freethink.com/opinion/studio-ghibli-chatgpt-creativity.

Post 3

At first, I used the prompt “Help me manage my time”, where Deepseek gave me a super interesting result combining different strategies that I can use:

  1. Clarity: Decide What Matters
    Before planning how to spend time, know what you’re optimizing for.

The 80/20 Rule: What 20% of your tasks give you 80% of your results (or fulfillment)? Identify those 1–3 things.

The 3–3–3 Method: Each day, plan:

3 hours on your most important project (deep work).

3 shorter urgent tasks (meetings, emails, admin).

3 maintenance tasks (exercise, chores, learning).

  1. Structure: Plan with Intention
    Don’t just react to what’s loudest. Use time blocking.

Time Blocking: Open your calendar. Assign specific blocks of time for deep work, shallow work, and breaks. Treat these blocks as non-negotiable appointments.

Batch Similar Tasks: Do all your calls in one block, all your email in one block. Context switching is a silent time-killer.

The Night Before: Spend 10 minutes each evening planning the next day. It saves hours of morning indecision.

  1. Execution: Protect Your Focus
    The best plan fails without execution discipline.

Single-Task: When working, focus on one thing. Close unrelated tabs, put your phone in another room.

Manage Energy, Not Just Time: Are you a morning person? Do your hardest work then. Low-energy after lunch? Schedule admin tasks or a walk.

Use a “Done” List: At the end of the day, list what you did accomplish. This builds momentum and prevents the feeling of “I got nothing done.”

  1. Recovery: Avoid Burnout
    Time management isn’t about maximizing every minute. It’s about sustainability.

Schedule Breaks: Use techniques like Pomodoro (25 min work, 5 min break) or simply ensure you take a 5–10 minute break every 90 minutes.

Define Shutdown: Have a clear end to your workday. A ritual (like closing your laptop or changing clothes) signals to your brain that work is done.

These are all useful, however no exact instruction of what “I should really do”, it even told me what matters the most but by not understanding what the most important thing is, because I did not provide other information but jsut to tell me how to manage my time.

Where the other prompt that I used, which is “I am a Junior college student, now I feel like my task is really difficult and I usually sleep around 12. Please help me with my time management when my average out of class time is 4pm.”, worked much better to give actual timetable to follow, tho, still not having other deeper intel because I did not provide any:

Summary of Your New Schedule:
Time Activity
4:00 – 4:15 Transition / Snack / Walk
4:15 – 5:30 HARDEST TASK (Phone away, deep focus)
5:30 – 7:00 Dinner + True Break (No work)
7:00 – 9:00 Easy tasks, review, organization
9:00 – 10:30 Shutdown routine / Wind down
10:30 In bed, lights low, no screens
11:00 Asleep

In this prompt, I used the suggestions from UT Guide:

To get your desired output, it is important to be specific about what that output should look like. When crafting a prompt, remember to try and incorporate these three things:

  • Who the author is
  • Who the audience is
  • What the purpose is
  • Example: “please generate a guide written from the perspective of a hiring manager communicating standards of professional attire in a business-casual setting to a group of newly hired student interns”

Which, I did include all these basic components and AI did provide a more specific answer.