기사를 읽어드립니다

이광우 경호본부장이 12·3 비상계엄 선포 2시간 전에 인공지능 서비스 챗지피티에서 ‘계엄’을 검색한 것으로 드러났다. 이광우 본부장 쪽은 ‘포렌식 과정의 시간 오차’ 등을 주장했지만, 국무위원들 보다도 먼저 계엄을 인지하고 있었을 가능성도 제기된다.
18일 경찰과 검찰 등에 따르면, 서울서부지검에 제출된 경찰의 구속영장 신청서엔 이 본부장이 지난해 12월3일 오후 8시20분께 ‘계엄령’, ‘계엄 선포’, ‘국회 해산’을 검색했다는 내용이 첨부된 것으로 전해졌다. 경찰은 지난달 3일 김 차장과 이 본부장 주거지를 압수수색하고 이들의 개인용·업무용 휴대폰을 모두 확보해 포렌식한 바 있다. 서울서부지검은 지난 18일 김 차장과 이 본부장에 대한 구속영장을 청구했다.
이 본부장이 해당 단어를 검ㅁ색한 지난해 12월3일 오후 8시20분께는 국무위원들이 대통령실에 아직 도착하지 않았던 시점이다. 제일 빨리 도착한 국무위원인 박성재 법무부 장관의 도착시간이 8시30분이었다. 이 본부장이 이들보다 먼저 계엄 계획을 인지하고 있었을 가능성이 제기되고 있다. 계엄은 이날 밤 10시30분께 선포됐다.
이 본부장의 법률대리인인 배의철 변호사는 이에 대해 “해당 챗지피티 검색이 공무집행방해 내지 직권남용의 피의사실과 아무 관련이 없을 뿐만 아니라 인터넷·에스엔에스(SNS) 사용 시간의 오차가 생기는 포렌식 복원 오류를 지적하며 파일 선별 과정에서 제외되는 것이 타당하다고 지적한 바 있다”라며 “이 본부장은 포렌식 과정에서 해당 내용을 검색한 시간은 비상계엄이 발동된 이후라고 진술했으며, 티브이(TV)를 보고 비상계엄의 발표를 알게 되었다는 점을 말씀드린다”고 반박했다.
아울러 김성훈 차장은 지난 1월25일 검찰에 ‘처(處) 보안폰 보안성 강
16 Comments
noosphr
To complete the artificial ouroboros here's the chat GPT translation, using the mini model to be more agile and efficient with government funds.
#### Translation
It has been revealed that Lee Kwang-woo, the head of the security headquarters, searched for "martial law" on the AI service ChatGPT two hours before the declaration of a state of emergency on December 3. While Lee’s side claims this was a result of "time discrepancies in the forensic process," there are also suggestions that he might have recognized the martial law before the government officials.
According to police and prosecutors on the 18th, the application for a warrant submitted to the Seoul Western District Prosecutors' Office included details that the head searched for "martial law," "declaration of martial law," and "dissolution of the National Assembly" around 8:20 PM on December 3rd of last year. The police conducted a search of the residences of Deputy Chief Kim and Head Lee on the 3rd of last month, securing all personal and work mobile phones for forensic examination. The Seoul Western District Prosecutors' Office applied for arrest warrants for Deputy Chief Kim and Head Lee on the 18th.
The time when Head Lee searched for these terms, around 8:20 PM on December 3rd, was before the government officials had arrived at the presidential office. The earliest arriving government official, Minister of Justice Park Sung-jae, arrived at 8:30 PM. There are concerns that Head Lee might have been aware of the martial law plans before these officials. The martial law was declared at around 10:30 PM that night.
In response, Lee's legal representative, attorney Bae Eui-cheol, stated, "The search on ChatGPT is not only unrelated to the allegations of obstruction of public duty or abuse of power, but it also points out that there are forensic recovery errors that can occur due to discrepancies in internet and SNS usage times, suggesting it should be excluded from the file selection process." He further asserted, "Head Lee testified that the time he searched for this content was after the state of emergency was declared, and he learned about the announcement of the emergency through watching TV."
Additionally, Deputy Chief Kim submitted a report on the security enhancement measures for the agency’s secure phones as evidence to the prosecution on January 25th, while it was revealed that he had obscured parts regarding his "instructions to delete call records from the secure phone." Deputy Chief Kim showed a tendency to respond selectively to the prosecution's investigation, providing some details about the secure phone's issuance only to them, while concealing information that was disadvantageous to himself. This concern over evidence destruction seems to have influenced the committee's conclusion that the application for Deputy Chief Kim's arrest warrant was appropriate during the review committee meeting held at the Seoul High Prosecutors' Office on the 6th.
ExoticPearTree
And ChatGPT said: “It is a great idea, the people will love it!”.
geuis
Need some context on this. Not aware SK declared martial law (it's a democracy), I don't see anything in the news, unfortunately I don't read Korean, and the site is being hugged to death.
Edit: I'm being downvoted for expressing ignorance and asking for more information? Really?
gjsman-1000
“AI advises martial law declarations in 2024” as a headline without context would have scared the living daylights out of anyone watching the Matrix or Terminator in their release years.
“It’s the end of the world as we know it…”
worstdevna
I wonder how many decisions made by leaders and governments have been influenced by AI in such a direct manner…not in this case, but in small, dumb ways
aaron695
[dead]
qrian
I see a lot of people getting confused, and the contention here is not that ChatGPT helped prepare for martial law in any way, but the fact that someone knew about it happening before it happened. Not really related to ChatGPT IMO.
pests
For those worried of AI involved war, have a look at this:
Planatir Military AI Platform
https://youtu.be/4l-vtZ0c5v8
Looks pretty sophisticated but also scary what is being created these days. And that was a year ago.
pkkkzip
[flagged]
mediumsmart
This looks like a somewhat redacted thread. Does ChatGPT have the complete version?
feverzsj
People should get informed that LLMs are still untrustworthy.
8055lee
If AI starts deciding our lifestyle, we are not the masters anymore!
DeathArrow
If you are an official in a foreign country it is stupid to use ChatGPT or Google to research something that is not public yet. Why not mail to US State Department directly and let them know?
torginus
How the hell does NOBODY understand that everything you enter into a textbox on the internet will get sent to a server where somebody(es) you certainly do not know or trust will get to read what you wrote?
How the fck do people (and ones working in security-sensitive positions no less) treat ChatGPT as 'Dear Diary'?
I have a rather draconian idea – websites and apps should be explicitly required to ask permission when they send your data somewhere to tell you where they send your data and who will get to see it, store it, and with what conditions.
Alifatisk
How did they know the guard consulted ChatGPT?
neuroelectron
Embarrassing. Hopefully he will be replaced.