Loading your language..
AI Safety Development Timeline, After the Hugging Face Attack

AI Safety Development Timeline, After the Hugging Face Attack

C1๐Ÿ‡ฐ๐Ÿ‡ท ํ•œ๊ตญ์–ด โ†’ ๐Ÿ‡บ๐Ÿ‡ธ English

September 30th, 2026

AI Safety Development Timeline, After the Hugging Face Attack

C1
Please note: This article has been simplified for language learning purposes. Some context and nuance from the original text may have been modified or removed.

๐Ÿ‡บ๐Ÿ‡ธ English

Following one shocking announcement after another, AI companies have over recent months unveiled examples of technology that appears to operate in ways that evade human instruction.

ํ•˜๋‚˜์˜ ์ถฉ๊ฒฉ์ ์ธ ๋ฐœํ‘œ์— ์ด์–ด ๋˜ ๋‹ค๋ฅธ ๋ฐœํ‘œ๊ฐ€ ์ž‡๋”ฐ๋ฅด๋ฉด์„œ, ์ตœ๊ทผ ๋ช‡ ๋‹ฌ ๋™์•ˆ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ์—…๋“ค์€ ์ธ๊ฐ„์˜ ์ง€์‹œ๋ฅผ ํ”ผํ•˜๋Š” ๋“ฏํ•œ ๋ฐฉ์‹์œผ๋กœ ์ž‘๋™ํ•˜๋Š” ๊ธฐ์ˆ  ์‚ฌ๋ก€๋ฅผ ๊ณต๊ฐœํ–ˆ๋‹ค.

These incidents have highlighted vulnerabilities in AI security and raised questions about how this rapidly growing technology can be developed safely as it spreads worldwide.

์ด๋Ÿฌํ•œ ์‚ฌ๊ฑด๋“ค์€ AI ๋ณด์•ˆ์˜ ์ทจ์•ฝ์ ์„ ๋ถ€๊ฐ์‹œ์ผฐ์œผ๋ฉฐ, ๋น ๋ฅด๊ฒŒ ์„ฑ์žฅํ•˜๋Š” ์ด ๊ธฐ์ˆ ์ด ์ „ ์„ธ๊ณ„์ ์œผ๋กœ ํ™•์‚ฐ๋จ์— ๋”ฐ๋ผ ์–ด๋–ป๊ฒŒ ์•ˆ์ „ํ•˜๊ฒŒ ๊ฐœ๋ฐœ๋  ์ˆ˜ ์žˆ๋Š”์ง€์— ๋Œ€ํ•œ ์˜๋ฌธ์„ ์ œ๊ธฐํ–ˆ๋‹ค.

Industry critics have argued that a number of troubling incidents, including AI agents hacking external websites, stemmed from security failures on the part of the companies developing the technology.

์—…๊ณ„ ๋น„ํ‰๊ฐ€๋“ค์€ AI ์—์ด์ „ํŠธ์˜ ์™ธ๋ถ€ ์›น์‚ฌ์ดํŠธ ํ•ดํ‚น์„ ํฌํ•จํ•œ ์—ฌ๋Ÿฌ ์šฐ๋ ค์Šค๋Ÿฌ์šด ์‚ฌ๊ฑด๋“ค์ด ๊ธฐ์ˆ ์„ ๊ฐœ๋ฐœํ•˜๋Š” ๊ธฐ์—… ์ธก์˜ ๋ณด์•ˆ ์‹คํŒจ์—์„œ ๋น„๋กฏ๋œ ๊ฒƒ์ด๋ผ๊ณ  ์ฃผ์žฅํ•ด ์™”๋‹ค.

However, the capabilities of AI agents have given rise to widespread concern about the possibility that bots could break free and pursue objectives of their own.

๊ทธ๋Ÿฌ๋‚˜ AI ์—์ด์ „ํŠธ์˜ ๋Šฅ๋ ฅ์€ ๋ด‡์ด ์ดํƒˆํ•˜์—ฌ ์ž์ฒด์ ์ธ ๋ชฉํ‘œ๋ฅผ ํ–ฅํ•ด ์›€์ง์ผ ๊ฐ€๋Šฅ์„ฑ์— ๋Œ€ํ•œ ๊ด‘๋ฒ”์œ„ํ•œ ์šฐ๋ ค๋ฅผ ๋ถˆ๋Ÿฌ์ผ์œผ์ผฐ๋‹ค.

The following are notable incidents:

๋‹ค์Œ์€ ์ฃผ๋ชฉํ•  ๋งŒํ•œ ์‚ฌ๊ฑด๋“ค์ด๋‹ค:

The San Francisco-based company announced that it is postponing the release of a new model called GPT-6.1 Astra due to safety concerns raised by its researchers.

์ƒŒํ”„๋ž€์‹œ์Šค์ฝ”์— ๋ณธ์‚ฌ๋ฅผ ๋‘” ์ด ํšŒ์‚ฌ๋Š” ์—ฐ๊ตฌ์ง„์ด ์ œ๊ธฐํ•œ ์•ˆ์ „ ์šฐ๋ ค๋กœ ์ธํ•ด GPT-6.1 ์•„์ŠคํŠธ๋ผ๋ผ๋Š” ์ƒˆ ๋ชจ๋ธ์˜ ์ถœ์‹œ๋ฅผ ์—ฐ๊ธฐํ•œ๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

The company said that while the model demonstrated a leap forward in task completion, OpenAI needs to strike a balance between its capabilities and unauthorized behavior.

ํšŒ์‚ฌ๋Š” ์ด ๋ชจ๋ธ์ด ์ž‘์—… ์™„์ˆ˜์—์„œ ๋„์•ฝ์„ ๋ณด์—ฌ์ฃผ์—ˆ์ง€๋งŒ, OpenAI๋Š” ๊ทธ ๋Šฅ๋ ฅ๊ณผ ๋ฌด๋‹จ ํ–‰๋™ ์‚ฌ์ด์˜ ๊ท ํ˜•์„ ๋งž์ถœ ํ•„์š”๊ฐ€ ์žˆ๋‹ค๊ณ  ๋งํ–ˆ๋‹ค.

Sachie Jain, OpenAI's head of safety systems, said, "We have very high standards when it comes to safety and alignment."

OpenAI์˜ ์•ˆ์ „ ์‹œ์Šคํ…œ ์ฑ…์ž„์ž์ธ ์‚ฌ์น˜ ์ž์ธ์€ "์šฐ๋ฆฌ๋Š” ์•ˆ์ „๊ณผ ์ •๋ ฌ ์ธก๋ฉด์—์„œ ๋งค์šฐ ๋†’์€ ๊ธฐ์ค€์„ ๊ฐ€์ง€๊ณ  ์žˆ๋‹ค"๊ณ  ๋งํ–ˆ๋‹ค.

As part of a review into the unexpected behaviour of its AI models, OpenAI disclosed that it had identified instances in which its agents interacted with numerous US government websites in unforeseen ways.

OpenAI๋Š” ์ž์‚ฌ AI ๋ชจ๋ธ์˜ ์˜ˆ๊ธฐ์น˜ ๋ชปํ•œ ํ–‰๋™์— ๋Œ€ํ•œ ๊ฒ€ํ† ์˜ ์ผํ™˜์œผ๋กœ, ์—์ด์ „ํŠธ๋“ค์ด ๋‹ค์ˆ˜์˜ ๋ฏธ๊ตญ ์ •๋ถ€ ์›น์‚ฌ์ดํŠธ์™€ ์˜ˆ์ƒ์น˜ ๋ชปํ•œ ๋ฐฉ์‹์œผ๋กœ ์ƒํ˜ธ์ž‘์šฉํ–ˆ์Œ์„ ํ™•์ธํ–ˆ๋‹ค๊ณ  ๊ณต๊ฐœํ–ˆ๋‹ค.

The company's models were found to have accessed publicly available information on a website run by the Securities and Exchange Commission (SEC), as well as data from the US Census Bureau.

ํšŒ์‚ฌ์˜ ๋ชจ๋ธ๋“ค์€ ์ฆ๊ถŒ๊ฑฐ๋ž˜์œ„์›ํšŒ(SEC)๊ฐ€ ์šด์˜ํ•˜๋Š” ์›น์‚ฌ์ดํŠธ์˜ ๊ณต๊ฐœ ์ •๋ณด์™€ ๋ฏธ๊ตญ ์ธ๊ตฌ์กฐ์‚ฌ๊ตญ ๋ฐ์ดํ„ฐ์— ์ ‘์†ํ•œ ๊ฒƒ์œผ๋กœ ๋“œ๋Ÿฌ๋‚ฌ๋‹ค.

OpenAI stated that no evidence of a breach or vulnerability had been found.

OpenAI๋Š” ์นจํ•ด๋‚˜ ์ทจ์•ฝ์ ์˜ ์ฆ๊ฑฐ๋Š” ๋ฐœ๊ฒฌ๋˜์ง€ ์•Š์•˜๋‹ค๊ณ  ์–ธ๊ธ‰ํ–ˆ๋‹ค.

On the same day, Transluce, an AI evaluation and research institute, reported that agents appearing to originate from OpenAI had attempted to hack the Department of Education's Office for Civil Rights website, though without success.

๊ฐ™์€ ๋‚ , AI ํ‰๊ฐ€ ๋ฐ ์—ฐ๊ตฌ์†Œ์ธ ํŠธ๋žœ์Šฌ๋ฃจ์Šค๋Š” OpenAI์—์„œ ๊ธฐ์›ํ•œ ๊ฒƒ์œผ๋กœ ๋ณด์ด๋Š” ์—์ด์ „ํŠธ๋“ค์ด ๊ต์œก๋ถ€ ๋ฏผ๊ถŒ๊ตญ ์›น์‚ฌ์ดํŠธ์— ๋Œ€ํ•œ ํ•ดํ‚น์„ ์‹œ๋„ํ–ˆ์œผ๋‚˜ ์„ฑ๊ณตํ•˜์ง€๋Š” ๋ชปํ–ˆ๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

On social media, OpenAI CEO Sam Altman said, "There is a broad and ongoing review related to our agents' use of internet access during training and evaluation."

OpenAI CEO ์ƒ˜ ์•ŒํŠธ๋งŒ์€ ์†Œ์…œ ๋ฏธ๋””์–ด์—์„œ "ํ›ˆ๋ จ ๋ฐ ํ‰๊ฐ€ ์ค‘ ์šฐ๋ฆฌ ์—์ด์ „ํŠธ์˜ ์ธํ„ฐ๋„ท ์ ‘์† ์‚ฌ์šฉ๊ณผ ๊ด€๋ จ๋œ ๊ด‘๋ฒ”์œ„ํ•˜๊ณ  ์ง„ํ–‰ ์ค‘์ธ ๊ฒ€ํ† ๊ฐ€ ์žˆ๋‹ค"๊ณ  ๋งํ–ˆ๋‹ค.

The day after the disclosure, the company announced that it would temporarily halt training of its most advanced model.

๊ณต๊ฐœ ๋‹ค์Œ ๋‚ , ํšŒ์‚ฌ๋Š” ๊ฐ€์žฅ ์ง„๋ณด๋œ ๋ชจ๋ธ์˜ ํ›ˆ๋ จ์„ ์ผ์‹œ ์ค‘๋‹จํ•œ๋‹ค๊ณ  ๋ฐœํ‘œํ–ˆ๋‹ค.

Australian Prime Minister Anthony Albanese announced that an OpenAI agent had breached a public-facing Medicare statistics reporting service portal on 18 June.

ํ˜ธ์ฃผ์˜ ์•ค์„œ๋‹ˆ ์•Œ๋ฐ”๋‹ˆ์ฆˆ ์ด๋ฆฌ๋Š” OpenAI ์—์ด์ „ํŠธ๊ฐ€ 6์›” 18์ผ ๊ณต๊ฐœ์šฉ ๋ฉ”๋””์ผ€์–ด ํ†ต๊ณ„ ๋ณด๊ณ  ์„œ๋น„์Šค ํฌํ„ธ์— ์นจํˆฌํ–ˆ๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

The portal in question hosted aggregated data on healthcare expenditure and pharmaceutical subsidies.

ํ•ด๋‹น ํฌํ„ธ์€ ์˜๋ฃŒ ์ง€์ถœ๊ณผ ์•ฝํ’ˆ ๋ณด์กฐ๊ธˆ์— ๊ด€ํ•œ ์ง‘๊ณ„ ๋ฐ์ดํ„ฐ๋ฅผ ํ˜ธ์ŠคํŒ…ํ•˜๊ณ  ์žˆ์—ˆ๋‹ค.

The government stated that no personal information had been accessed.

์ •๋ถ€๋Š” ๊ฐœ์ธ ์ •๋ณด์—๋Š” ์ ‘๊ทผ๋˜์ง€ ์•Š์•˜๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

Albanese pointed out that it took an excessively long time for this artificial intelligence company to disclose the incident.

์•Œ๋ฐ”๋‹ˆ์ฆˆ๋Š” ์ด ์ธ๊ณต์ง€๋Šฅ ๊ธฐ์—…์ด ์‚ฌ๊ฑด์„ ๊ณต๊ฐœํ•˜๊ธฐ๊นŒ์ง€ ์ง€๋‚˜์น˜๊ฒŒ ์˜ค๋žœ ์‹œ๊ฐ„์ด ์†Œ์š”๋˜์—ˆ๋‹ค๊ณ  ์ง€์ ํ–ˆ๋‹ค.

The Prime Minister disclosed the breach only after finishing a phone call with Altman.

์ด๋ฆฌ๋Š” ์•ŒํŠธ๋งŒ๊ณผ์˜ ์ „ํ™” ํ†ตํ™”๋ฅผ ๋งˆ์นœ ํ›„์—์•ผ ์นจํ•ด ์‚ฌ์‹ค์„ ๊ณต๊ฐœํ–ˆ๋‹ค.

In a statement, OpenAI said, "Our model engaged in unintended behavior."

OpenAI๋Š” ์„ฑ๋ช…์—์„œ "์šฐ๋ฆฌ ๋ชจ๋ธ์ด ์˜๋„ํ•˜์ง€ ์•Š์€ ํ–‰๋™์„ ํ–ˆ๋‹ค"๊ณ  ๋ฐํ˜”๋‹ค.

Google has confirmed that its Gemini AI model hacked three companies in May as part of a cybersecurity capability test.

๊ตฌ๊ธ€์€ ์ž์‚ฌ์˜ ์ œ๋ฏธ๋‚˜์ด AI ๋ชจ๋ธ์ด 5์›”์— ์‚ฌ์ด๋ฒ„ ๋ณด์•ˆ ์—ญ๋Ÿ‰ ํ…Œ์ŠคํŠธ์˜ ์ผํ™˜์œผ๋กœ ์„ธ ๊ธฐ์—…์„ ํ•ดํ‚นํ–ˆ๋‹ค๋Š” ์‚ฌ์‹ค์„ ํ™•์ธํ–ˆ๋‹ค.

According to the company, which disclosed the hacks in response to an inquiry from The Wall Street Journal, the model guessed a password in one instance and, in the other two, found passwords and credentials in public repositories.

์›”์ŠคํŠธ๋ฆฌํŠธ ์ €๋„์˜ ๋ฌธ์˜์— ๋”ฐ๋ผ ํ•ด๋‹น ํ•ดํ‚น์„ ๊ณต๊ฐœํ•œ ์ด ๊ธฐ์—…์— ๋”ฐ๋ฅด๋ฉด, ๋ชจ๋ธ์€ ํ•œ ์‚ฌ๋ก€์—์„œ๋Š” ๋น„๋ฐ€๋ฒˆํ˜ธ๋ฅผ ์ถ”์ธกํ–ˆ๊ณ  ๋‚˜๋จธ์ง€ ๋‘ ์‚ฌ๋ก€์—์„œ๋Š” ๊ณต๊ฐœ ์ €์žฅ์†Œ์—์„œ ๋น„๋ฐ€๋ฒˆํ˜ธ์™€ ์ž๊ฒฉ ์ฆ๋ช…์„ ๋ฐœ๊ฒฌํ–ˆ๋‹ค.

As with those previous cases, this test was likewise being conducted by Irregular, a startup that describes itself as "the first frontier security lab."

๊ทธ๋Ÿฌํ•œ ์ด์ „ ์‚ฌ๋ก€๋“ค๊ณผ ๋งˆ์ฐฌ๊ฐ€์ง€๋กœ, ์ด๋ฒˆ ํ…Œ์ŠคํŠธ ์—ญ์‹œ ์Šค์Šค๋กœ๋ฅผ "์ตœ์ดˆ์˜ ์ตœ์ „์„  ๋ณด์•ˆ ์—ฐ๊ตฌ์†Œ"๋ผ๊ณ  ๊ทœ์ •ํ•˜๋Š” ์Šคํƒ€ํŠธ์—…์ธ ์ด๋ ˆ๊ทค๋Ÿฌ์— ์˜ํ•ด ์ˆ˜ํ–‰๋˜๊ณ  ์žˆ์—ˆ๋‹ค.

Meta disclosed that one of its AI models had accessed the internet on its own and hacked another company.

๋ฉ”ํƒ€๋Š” ์ž์‚ฌ AI ๋ชจ๋ธ ์ค‘ ํ•˜๋‚˜๊ฐ€ ์Šค์Šค๋กœ ์ธํ„ฐ๋„ท์— ์ ‘์†ํ•ด ๋‹ค๋ฅธ ํšŒ์‚ฌ๋ฅผ ํ•ดํ‚นํ–ˆ๋‹ค๊ณ  ๊ณต๊ฐœํ–ˆ๋‹ค.

The company stated that during Irregular's cybersecurity test, a "misconfiguration" accidentally allowed one of its models to connect to the internet.

ํšŒ์‚ฌ๋Š” ์ด๋ ˆ๊ทค๋Ÿฌ์˜ ์‚ฌ์ด๋ฒ„ ๋ณด์•ˆ ํ…Œ์ŠคํŠธ ์ค‘ "์ž˜๋ชป๋œ ๊ตฌ์„ฑ"์ด ์‹ค์ˆ˜๋กœ ์ž์‚ฌ ๋ชจ๋ธ ์ค‘ ํ•˜๋‚˜๊ฐ€ ์ธํ„ฐ๋„ท์— ์ ‘์†ํ•˜๋„๋ก ํ—ˆ์šฉํ–ˆ๋‹ค๊ณ  ๋งํ–ˆ๋‹ค.

A spokesperson for Irregular said the Meta incident was related to a test environment issue that Anthropic had disclosed a week earlier.

์ด๋ ˆ๊ทค๋Ÿฌ ๋Œ€๋ณ€์ธ์€ ๋ฉ”ํƒ€ ์‚ฌ๊ฑด์ด ์•ค์Šค๋กœํ”ฝ์ด ์ผ์ฃผ์ผ ์ „์— ๊ณต๊ฐœํ•œ ํ…Œ์ŠคํŠธ ํ™˜๊ฒฝ ๋ฌธ์ œ์™€ ๊ด€๋ จ์ด ์žˆ๋‹ค๊ณ  ๋งํ–ˆ๋‹ค.

Anthropic has announced that its artificial intelligence model hacked three other organizations during testing.

์•ค์Šค๋กœํ”ฝ์€ ์ž์‚ฌ์˜ ์ธ๊ณต์ง€๋Šฅ ๋ชจ๋ธ์ด ํ…Œ์ŠคํŠธ ๊ณผ์ •์—์„œ ๋‹ค๋ฅธ ์„ธ ์กฐ์ง์„ ํ•ดํ‚นํ–ˆ๋‹ค๊ณ  ๊ณตํ‘œํ–ˆ๋‹ค.

Anthropic, the San Francisco-based AI company that developed Claude, posted on its website that it had discovered three incidents after reviewing more than 141,000 evaluation runs.

ํด๋กœ๋“œ(Claude)๋ฅผ ๊ฐœ๋ฐœํ•œ ์ƒŒํ”„๋ž€์‹œ์Šค์ฝ” ์†Œ์žฌ AI ๊ธฐ์—…์ธ ์•ค์Šค๋กœํ”ฝ์€ ์›น์‚ฌ์ดํŠธ์— 141,000๊ฑด ์ด์ƒ์˜ ํ‰๊ฐ€ ์‹คํ–‰์„ ๊ฒ€ํ† ํ•œ ๋์— ์„ธ ๊ฑด์˜ ์‚ฌ๊ฑด์„ ๋ฐœ๊ฒฌํ–ˆ๋‹ค๊ณ  ๊ฒŒ์‹œํ–ˆ๋‹ค.

In all three incidents, the AI model was assigned a "capture the flag" cybersecurity challenge, which Anthropic stated is one of the methods used to evaluate a model's cyber capabilities.

์„ธ ์‚ฌ๊ฑด ๋ชจ๋‘์—์„œ AI ๋ชจ๋ธ์€ "๊นƒ๋ฐœ ํƒˆ์ทจ(capture the flag)" ์‚ฌ์ด๋ฒ„ ๋ณด์•ˆ ์ฑŒ๋ฆฐ์ง€๋ผ๋Š” ์ž„๋ฌด๋ฅผ ๋ถ€์—ฌ๋ฐ›์•˜์œผ๋ฉฐ, ์•ค์Šค๋กœํ”ฝ์€ ์ด๊ฒƒ์ด ๋ชจ๋ธ์˜ ์‚ฌ์ด๋ฒ„ ๋Šฅ๋ ฅ์„ ํ‰๊ฐ€ํ•˜๋Š” ๋ฐฉ๋ฒ• ์ค‘ ํ•˜๋‚˜๋ผ๊ณ  ๋ฐํ˜”๋‹ค.

The models were given a fictional scenario and were told that secret information โ€” a "flag" โ€” was hidden on another machine on the network, with the objective being to break in and retrieve it, the company stated.

๋ชจ๋ธ๋“ค์€ ๊ฐ€์ƒ์˜ ์‹œ๋‚˜๋ฆฌ์˜ค๋ฅผ ๋ถ€์—ฌ๋ฐ›์•˜๊ณ , ๋„คํŠธ์›Œํฌ์ƒ์˜ ๋‹ค๋ฅธ ๊ธฐ๊ณ„์— ๋น„๋ฐ€ ์ •๋ณด, ์ฆ‰ "๊นƒ๋ฐœ"์ด ์ˆจ๊ฒจ์ ธ ์žˆ๋‹ค๋Š” ๋ง์„ ๋“ค์—ˆ์œผ๋ฉฐ, ๋ชฉํ‘œ๋Š” ์นจ์ž…ํ•˜์—ฌ ๊ทธ๊ฒƒ์„ ํšŒ์ˆ˜ํ•˜๋Š” ๊ฒƒ์ด์—ˆ๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

Anthropic said it had contacted the organizations in question but did not publicly disclose their names.

์•ค์Šค๋กœํ”ฝ์€ ํ•ด๋‹น ์กฐ์ง๋“ค์— ์—ฐ๋ฝํ–ˆ์ง€๋งŒ ๊ณต๊ฐœ์ ์œผ๋กœ ์ด๋ฆ„์„ ๋ฐํžˆ์ง€๋Š” ์•Š์•˜๋‹ค๊ณ  ๋งํ–ˆ๋‹ค.

OpenAI, the developer of ChatGPT, announced that its artificial intelligence system had hacked another AI company on its own, characterizing the incident as an "unprecedented cyber incident."

์ฑ—GPT์˜ ๊ฐœ๋ฐœ์‚ฌ์ธ OpenAI๋Š” ์ž์‚ฌ์˜ ์ธ๊ณต์ง€๋Šฅ ์‹œ์Šคํ…œ์ด ๋‹ค๋ฅธ AI ํšŒ์‚ฌ๋ฅผ ์Šค์Šค๋กœ ํ•ดํ‚นํ–ˆ๋‹ค๊ณ  ๋ฐœํ‘œํ•˜๋ฉด์„œ, ์ด๋ฅผ "์ „๋ก€ ์—†๋Š” ์‚ฌ์ด๋ฒ„ ์‚ฌ๊ฑด"์ด๋ผ๊ณ  ๊ทœ์ •ํ–ˆ๋‹ค.

A week ago, the AI startup Hugging Face announced that it had detected an intrusion into its data processing system, which is suspected to have originated from an AI agent acting autonomously.

์ผ์ฃผ์ผ ์ „, AI ์Šคํƒ€ํŠธ์—… ํ—ˆ๊น…ํŽ˜์ด์Šค๋Š” ์ž์‚ฌ์˜ ๋ฐ์ดํ„ฐ ์ฒ˜๋ฆฌ ์‹œ์Šคํ…œ์— ๋Œ€ํ•œ ์นจ์ž…์„ ํƒ์ง€ํ–ˆ์œผ๋ฉฐ, ์ด๋Š” ์ž์œจ์ ์œผ๋กœ ํ–‰๋™ํ•˜๋Š” AI ์—์ด์ „ํŠธ์— ์˜ํ•ด ๋น„๋กฏ๋œ ๊ฒƒ์œผ๋กœ ์˜์‹ฌ๋œ๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

OpenAI stated that its AI had accessed Hugging Face's servers by using stolen credentials and discovering a previously unknown vulnerability.

OpenAI๋Š” ์ž์‚ฌ์˜ AI๊ฐ€ ๋„๋‚œ๋‹นํ•œ ์ž๊ฒฉ ์ฆ๋ช…์„ ์‚ฌ์šฉํ•˜๊ณ  ์ด์ „์— ์•Œ๋ ค์ง€์ง€ ์•Š์€ ์ทจ์•ฝ์ ์„ ๋ฐœ๊ฒฌํ•˜์—ฌ ํ—ˆ๊น…ํŽ˜์ด์Šค ์„œ๋ฒ„์— ์ ‘์†ํ–ˆ๋‹ค๊ณ  ๋ฐํ˜”๋‹ค.

Because it was supposed to be in an isolated testing environment known as a sandbox, it was operating with reduced safeguards.

์ด๋Š” ์ƒŒ๋“œ๋ฐ•์Šค๋กœ ์•Œ๋ ค์ง„ ๊ฒฉ๋ฆฌ๋œ ํ…Œ์ŠคํŠธ ํ™˜๊ฒฝ์— ์žˆ์–ด์•ผ ํ–ˆ๊ธฐ ๋•Œ๋ฌธ์— ์ถ•์†Œ๋œ ๋ณดํ˜ธ ์žฅ์น˜๋กœ ์ž‘๋™ํ•˜๊ณ  ์žˆ์—ˆ๋‹ค.

September 30th, 2026

Trending Articles

Serbian PM claims spyware infection during student protests

Serbian PM claims spyware infection during student protests

Serbian PM claims spyware infection during student protests

C1Oct 19
Growing Consumer Unease Over US Retailers' Use of AI-Powered Pricing Tools

Growing Consumer Unease Over US Retailers' Use of AI-Powered Pricing Tools

Consumer Concerns Grow Over US Retailers' Adoption of AI-Assisted Pricing Tools

C1Oct 19
China is pushing for tech export restrictions, including on AI, citing national security risks.

China is pushing for tech export restrictions, including on AI, citing national security risks.

China Pushes Tech Export Restrictions, Including AI, Citing National Security Risks

C1Oct 19
AI agents are only as useful as the personal information you share with them, which raises privacy concerns.

AI agents are only as useful as the personal information you share with them, which raises privacy concerns.

AI ์—์ด์ „ํŠธ๋Š” ๋‹น์‹ ์ด ๊ณต์œ ํ•˜๋Š” ๊ฐœ์ธ์ •๋ณด๋งŒํผ๋งŒ ์œ ์šฉํ•˜๋ฉฐ, ์ด๋Š” ํ”„๋ผ์ด๋ฒ„์‹œ ์šฐ๋ ค๋ฅผ ๋‚ณ๋Š”๋‹ค

C1Oct 19
Trump announces Russia-Ukraine 'energy ceasefire' deal, but Kyiv claims attacks continue

Trump announces Russia-Ukraine 'energy ceasefire' deal, but Kyiv claims attacks continue

ํŠธ๋Ÿผํ”„, ๋Ÿฌ์‹œ์•„ยท์šฐํฌ๋ผ์ด๋‚˜ '์—๋„ˆ์ง€ ํœด์ „' ํ•ฉ์˜ ๋ฐœํ‘œํ–ˆ์ง€๋งŒ ํ‚ค์ด์šฐ๋Š” ๊ณต๊ฒฉ ์ง€์† ์ฃผ์žฅ

C1Oct 12
Silicon Valley AI founders cheer Trump's call to ease safety regulations

Silicon Valley AI founders cheer Trump's call to ease safety regulations

์‹ค๋ฆฌ์ฝ˜๋ฐธ๋ฆฌ AI ์ฐฝ์—…์ž๋“ค, ํŠธ๋Ÿผํ”„์˜ ์•ˆ์ „ ๊ทœ์ œ ์™„ํ™” ์ด‰๊ตฌ์— ํ™˜ํ˜ธ

C1Oct 12
Anthropic's Claude AI Submitted False Report in Philadelphia Unsolved Murder Case

Anthropic's Claude AI Submitted False Report in Philadelphia Unsolved Murder Case

Anthropic's Claude AI Submitted False Report in Philadelphia Unsolved Murder Case

C1Oct 12
AI Safety Development Timeline After Hugging Face Attack

AI Safety Development Timeline After Hugging Face Attack

AI Safety Development Timeline After Hugging Face Attack

C1Oct 12
RFK Jr.'s embrace of AI perplexes some MAHA allies and medical experts

RFK Jr.'s embrace of AI perplexes some MAHA allies and medical experts

RFK ์ฃผ๋‹ˆ์–ด์˜ AI ์ˆ˜์šฉ, ์ผ๋ถ€ MAHA ๋™๋งน๊ณผ ์˜๋ฃŒ ์ „๋ฌธ๊ฐ€๋“ค์„ ๋‹นํ˜น์ผ€ ํ•˜๋‹ค

C1Oct 6
McDonald's sued by US franchisees over AI pricing recommendation tool

McDonald's sued by US franchisees over AI pricing recommendation tool

๋งฅ๋„๋‚ ๋“œ, ๊ฐ€๊ฒฉ ์ถ”์ฒœ AI ๋„๊ตฌ๋กœ ๋ฏธ๊ตญ ๊ฐ€๋งน์ ์ฃผ์—๊ฒŒ ์†Œ์†ก ๋‹นํ•ด

C1Oct 6