Re: [Vo]:ChatGPT translation of Takahashi's comment about plasma fusion

[Jed Rothwell]

Alain Sepeda  wrote:

I've seen the demo of an interesting tool, ResearchGPT, to have a
> discussion with a scientific paper.
>

I would like to add this kind of capability to LENR-CANR.org. I would like
a local copy of ChatGPT to read all of the papers, exclude all other data
sources on the internet, and answer questions about the contents of
LENR-CANR.org. It would work like the Google Custom Search box at the top
of every page at LENR-CANR.org, which says:

This Google™ Custom Search finds files at LENR-CANR only


(You can accomplish the same thing with a Google search beginning
"site:LENR-CANR.org . . .")

I do not know if that feature will be offered, but if it is, I will sign up
for it. I expect Microsoft will offer local copies of ChatGPT to big
corporations for their customer support operations, and to big universities
for their admissions, student info., course offerings and so on. If they
charge money to me for using it at LENR-CANR.org, I expect it will be a
nominal amount, because LENR-CANR.org traffic is low compared to big
corporate sites.

Google better hop to it and get their ChatBot working, or Microsoft will
eat their lunch.

Re: [Vo]:ChatGPT translation of Takahashi's comment about plasma fusion

[Alain Sepeda]

I've seen the demo of an interesting tool, ResearchGPT, to have a
discussion with a scientific paper.

"

Introducing researchGPT – An open-source research assistant that allows you
to have a conversation with a research paper or any pdf. Repo linked the
comments.

"
Maybe some researchers could have access to the beta version.

https://www.reddit.com/r/OpenAI/comments/117mre5/introducing_researchgpt_an_opensource_research/



Le mar. 28 févr. 2023 à 16:20, Jed Rothwell  a
écrit :

> Akito Takahashi commented on the laser fusion experiment that was in the
> news some months ago. I translated his comments. Then I had ChatGPT and
> translate.google.com translate them. ChatGPT is almost as good as a
> human. It almost could have fooled me, except for one or two words.
> translate.google.com made several mistakes. You can see it is a computer
> translation. I think it got the basic meaning correct.
>
> ChatGPT made a strange error. It left out a parenthetical comment. I told
> it about that, and it generated a new version. It also said "neutral
> neutron" which a human would not say.
>
> Here is the ChatGPT translation. The part it left out in the first version
> is underlined:
>
> This is a laser implosion D-T fusion reaction with a COP of 1.2. As an
> experiment, it is groundbreaking because it exceeded the energy break even
> (COP=1). However, in order to implement it in society, it requires a
> technology that can repeatedly shoot multiple high-intensity beams (36?) of
> laser pulses with a COP>5 and synchronize the injection of DT solid
> pellets. Without this, it cannot be a power plant. The 14MeV neutral
> neutron from the DT reaction *(my old specialty)*, as well as technology
> to convert it into heat energy using a Li blanket, radiation damage and
> countermeasures due to neutral materials, and the proliferation of T
> (tritium), which does not occur naturally, are required. It will take
> decades to become a commercial reactor."
>
> (In short, the current situation is that there is exaggerated advertising
> for the budget acquisition of a large number of troops behind it.)
>
> Here is my correction to ChatGPT after the first version. This shows you
> can have a bilingual conversation:
>
> You left out this part: （私の昔の専門）, which modifies 14MeV中性子
>
>
> Here is the translate.google.com version:
>
> One shot laser implosion D-T pulse fusion reaction at COP=1.2. As an
> experiment, it is epoch-making because it exceeded the energy break even
> (COP=1). For social implementation, high-intensity multiple-beam (36 ?)
> laser-pulses can be fired repeatedly at 10 shots per second with COP>5, and
> without the technology to synchronously inject DT solid pellets at the
> target, power generation reactors not. Technological development is
> required for the conversion of 14 MeV neutrons (my old specialty) in the DT
> reaction into thermal energy with a Li blanket, countermeasures against
> activation and damage of materials by neutrons, and growth of unnatural T
> (tritium). It will be several decades before it becomes a commercial
> furnace.
>
> (In short, the current situation is behind the hype to obtain a budget for
> the lives of a large number of troops.)
>
> My version:
>
> This is a laser implosion D-T fusion reaction with a COP of 1.2. As an
> experiment, it is groundbreaking, because it produced energy exceeding
> break even (COP=1). However, in order to implement this in a practical,
> commercial generator, we will need technology that can repeatedly shoot
> multiple high-intensity laser pulse beams (perhaps 36?) with a COP>5,
> synchronized with the injection of DT solid pellets. Without such
> technology, this cannot be used in a power plant. Dealing with 14 MeV
> neutron radiation from the DT reaction (my old specialty), will require
> extensive new technology, first to convert the reaction into heat energy
> using a Li blanket, and to deal with damage from neutron activation of the
> blanket material, and the proliferation of T (tritium), a radioactive
> material not found in nature. It will take decades to develop such
> technology to make this into a commercial reactor.
>
> (In short, the current situation is that this is exaggerated public
> relations puffery to obtain the budget from Congress, which must support
> the large number of researchers needed for this project.)
>
>

[Vo]:ChatGPT translation of Takahashi's comment about plasma fusion

[Jed Rothwell]

Akito Takahashi commented on the laser fusion experiment that was in the
news some months ago. I translated his comments. Then I had ChatGPT and
translate.google.com translate them. ChatGPT is almost as good as a human.
It almost could have fooled me, except for one or two words.
translate.google.com made several mistakes. You can see it is a computer
translation. I think it got the basic meaning correct.

ChatGPT made a strange error. It left out a parenthetical comment. I told
it about that, and it generated a new version. It also said "neutral
neutron" which a human would not say.

Here is the ChatGPT translation. The part it left out in the first version
is underlined:

This is a laser implosion D-T fusion reaction with a COP of 1.2. As an
experiment, it is groundbreaking because it exceeded the energy break even
(COP=1). However, in order to implement it in society, it requires a
technology that can repeatedly shoot multiple high-intensity beams (36?) of
laser pulses with a COP>5 and synchronize the injection of DT solid
pellets. Without this, it cannot be a power plant. The 14MeV neutral
neutron from the DT reaction *(my old specialty)*, as well as technology to
convert it into heat energy using a Li blanket, radiation damage and
countermeasures due to neutral materials, and the proliferation of T
(tritium), which does not occur naturally, are required. It will take
decades to become a commercial reactor."

(In short, the current situation is that there is exaggerated advertising
for the budget acquisition of a large number of troops behind it.)

Here is my correction to ChatGPT after the first version. This shows you
can have a bilingual conversation:

You left out this part: （私の昔の専門）, which modifies 14MeV中性子


Here is the translate.google.com version:

One shot laser implosion D-T pulse fusion reaction at COP=1.2. As an
experiment, it is epoch-making because it exceeded the energy break even
(COP=1). For social implementation, high-intensity multiple-beam (36 ?)
laser-pulses can be fired repeatedly at 10 shots per second with COP>5, and
without the technology to synchronously inject DT solid pellets at the
target, power generation reactors not. Technological development is
required for the conversion of 14 MeV neutrons (my old specialty) in the DT
reaction into thermal energy with a Li blanket, countermeasures against
activation and damage of materials by neutrons, and growth of unnatural T
(tritium). It will be several decades before it becomes a commercial
furnace.

(In short, the current situation is behind the hype to obtain a budget for
the lives of a large number of troops.)

My version:

This is a laser implosion D-T fusion reaction with a COP of 1.2. As an
experiment, it is groundbreaking, because it produced energy exceeding
break even (COP=1). However, in order to implement this in a practical,
commercial generator, we will need technology that can repeatedly shoot
multiple high-intensity laser pulse beams (perhaps 36?) with a COP>5,
synchronized with the injection of DT solid pellets. Without such
technology, this cannot be used in a power plant. Dealing with 14 MeV
neutron radiation from the DT reaction (my old specialty), will require
extensive new technology, first to convert the reaction into heat energy
using a Li blanket, and to deal with damage from neutron activation of the
blanket material, and the proliferation of T (tritium), a radioactive
material not found in nature. It will take decades to develop such
technology to make this into a commercial reactor.

(In short, the current situation is that this is exaggerated public
relations puffery to obtain the budget from Congress, which must support
the large number of researchers needed for this project.)