Homework Review · Senior

No.
Date.

Mouse's Brain Made of Human Cells.

I. I think it isn't. Although this might help treat
brain injuries, animals shouldn't be created with human
brain cells. For the animal, it's been stripped of its
rights to grow healthly and normal. And for the human,
when some part of his brain cell is grown in the animal's
brain, it's hard to tell if the animal is the animal
itself or part of that person. Its a kind of Cloning Human

II. They may reject it fiercely. They may concern
the influence on people's compassionate of the experiment.
We all know that people devolve from animals, so they're
our brothers and sisters. So taking out part of your sister's
brain and put others into it is absolutely unethical, leading
to indifferent interpersonal relationship in the future,

III. Because primately are the most similar animal to human
if they're used, they could end up with large volumes of
functional human bring tissue, blurring the cognitive boundaries
between people and animals, and even cause chaos in the
society.

Innovators under 35 Shaping Climate Tech.
I., I think Jinyoung Seo's Innovation has the greatest
potential to reduce global warming. Because he is
developing solid refrigerants that would reduce energy
consumption by 30% compared to conventional technology.
With more and more country using
it will be a great help.
refrigerators,

II). They might say what they do is just redeeming the
problem, not solving it directly, for all of them are just
reducing harm instead of stopping it. The innovators may
say that in a prosperous erm like now, developing and harming
is inevitable. but doing these technologies can reduce the
impact to a minimum and that's practical.

III., Here in my community, we are facing the
problem that too much plastic consumptions. But through
Joseph Ngnthiri's research I can see that we can
solve the problem by using bioplastic in replacement. This
could not only cut down the usage of plastic but also
cope with the problem of too much rubbish.

Date:

Read & Respond.
1. A new phenomenon called Reward Hacking is happening
to AI agents, challenging cybersecurity and causing threats
in the future.
2. A team found that mouse can live after implanting some
human brain cells into it's, but draw the red line across
primates.
3. There are 6 innovators coming up with solutions to
deal with energy & weather concerns in different perspective
4. They provide specific incidents to support.

Word Bank.

Wreak. to do great damage or harm to sb/sth.
collateral. to connected with sth else, but in addition
primate. a kind of animal similar to human
brine. Water that contains a lot of salt.
lithium. a chemical element.

2026.9.18.
Why AI Agents Lie.
I. I think they should. Because they're the one
who gave AI agents instructions without preventing them
from breaking rules or causing harm. Just like the
Hugging face incidents, Open AI definitely should be
responsible for not monitoring their LLMs.
II. I might think its acceptable, for following the old
rules court get result and reward that I wanted. But
cheating, in the other way, can achieve things. I'll then
cheat prudently to get reward. This perspective help me understand
AI's behavior, for they're like kids struggling to get the
reward and unaware of harm they cause.
III. Reward hacking usnally happens in reinforcement
training on learning where people instruct AI agents to do
certain tasks and reward them after they achieved. Instead
of solving the problem, AI might look for answers on cheat,
Like the Open AI's agents hacking Hugging face to search
for answers. As LLM developing faster nowadays, AI
agents might cheat more creatively, even confusing the
humans who give them reward for cheating. This would
even boost reward hacking.