"낙서장"의 두 판 사이의 차이
라이언의 꿀팁백과
| 1번째 줄: | 1번째 줄: | ||
<nowiki>#</nowiki> Part I - Choose five of the words/phrases and create new sentences. | |||
--- | |||
1. | 1. fair use | ||
→ The fair use doctrine enables the Internet to be abundant with data. | |||
--- | |||
2. | 2. credibility | ||
→ We need to strive to create safe and reliable AI models. In other words, we aim to ensure the credibility of the AI models we develop. | |||
--- | |||
3. on edge | |||
→ The kids in the family were on edge when their parents yelled at each other. | |||
--- | |||
4. on the back of | |||
→ I believe that God is on the back of those who are good and trustful in Him. | |||
--- | |||
5. hammer out something | |||
→ After hours of intense discussion, the committee was finally able to hammer out the details of the requirements of the project. | |||
--- | |||
<nowiki>#</nowiki> Part II Questions - Please write down the answers in the answer box below. | |||
--- | |||
1. How does ChatGPT gather information, and what has led to the lawsuit between The New York Times and OpenAI? | |||
→ ChatGPT collects a wide range of data from various sources, including websites on the internet and documents. | |||
The New York Times filed a lawsuit against OpenAI, alleging copyright infringement and claiming that OpenAI has harmed their reputation by spreading misinformation. | |||
--- | |||
2. What are the key arguments presented by both The New York Times and OpenAI in this copyright infringement case? | |||
→ Firstly, the licensing fee is a key point. The article mentions that The New York Times had attempted to negotiate a licensing deal for using their articles before filing the lawsuit. | |||
Secondly, the fair use doctrine, which permits the use of copyrighted material without permission under certain circumstances, is another crucial aspect. The outcome of this lawsuit could profoundly impact the way we handle data. | |||
--- | |||
3. According to the article, what potential consequences or changes might this lawsuit bring to the digital publishing industry and AI technology? | |||
→ Speaking of AI technology, the outcomes could be seriously disruptive if their use of data does not constitute fair use. As for the digital publishing industry, it could create another cash cow through licensing fees. | |||
--- | |||
4. Do you believe that AI entities like ChatGPT should be held accountable for using copyrighted material without explicit permission, or do you support the fair use argument presented by OpenAI? Why? | |||
→ Data are the lifeblood of AI technology, as AI models can only be trained with vast amounts of data. However, this data is not free. To produce it, companies must invest resources. Therefore, I believe a revised version of the fair use doctrine is necessary. For instance, what if the law only allowed companies to use public materials that were published more than two years ago? | |||
--- | |||
5. How might this lawsuit between The New York Times and OpenAI impact the future development and regulation of AI technologies in various industries, considering their use of copyrighted material? | |||
→ Policymakers should revise the current copyright laws, as they are not suitable for new technologies like AI unless updated. | |||
--- | |||
<nowiki>#</nowiki> | <nowiki>#</nowiki> | ||
Some food for thoughts: | |||
1. Case Tracker: Artificial Intelligence, Copyrights and Class Actions (New York Times v. Microsoft) (<nowiki>https://www.bakerlaw.com/new-york-times-v-microsoft/</nowiki>) | |||
2. "Nightshade", defensive tool for artists against AI art generators, poisons AI training models that scrape their work without permission (<nowiki>https://amt-lab.org/reviews/2023/11/nightshade-a-defensive-tool-for-artists-against-ai-art-generators</nowiki>) | |||
--- | |||
2024년 1월 29일 (월) 09:58 판
# Part I - Choose five of the words/phrases and create new sentences.
---
1. fair use
→ The fair use doctrine enables the Internet to be abundant with data.
---
2. credibility
→ We need to strive to create safe and reliable AI models. In other words, we aim to ensure the credibility of the AI models we develop.
---
3. on edge
→ The kids in the family were on edge when their parents yelled at each other.
---
4. on the back of
→ I believe that God is on the back of those who are good and trustful in Him.
---
5. hammer out something
→ After hours of intense discussion, the committee was finally able to hammer out the details of the requirements of the project.
---
# Part II Questions - Please write down the answers in the answer box below.
---
1. How does ChatGPT gather information, and what has led to the lawsuit between The New York Times and OpenAI?
→ ChatGPT collects a wide range of data from various sources, including websites on the internet and documents.
The New York Times filed a lawsuit against OpenAI, alleging copyright infringement and claiming that OpenAI has harmed their reputation by spreading misinformation.
---
2. What are the key arguments presented by both The New York Times and OpenAI in this copyright infringement case?
→ Firstly, the licensing fee is a key point. The article mentions that The New York Times had attempted to negotiate a licensing deal for using their articles before filing the lawsuit.
Secondly, the fair use doctrine, which permits the use of copyrighted material without permission under certain circumstances, is another crucial aspect. The outcome of this lawsuit could profoundly impact the way we handle data.
---
3. According to the article, what potential consequences or changes might this lawsuit bring to the digital publishing industry and AI technology?
→ Speaking of AI technology, the outcomes could be seriously disruptive if their use of data does not constitute fair use. As for the digital publishing industry, it could create another cash cow through licensing fees.
---
4. Do you believe that AI entities like ChatGPT should be held accountable for using copyrighted material without explicit permission, or do you support the fair use argument presented by OpenAI? Why?
→ Data are the lifeblood of AI technology, as AI models can only be trained with vast amounts of data. However, this data is not free. To produce it, companies must invest resources. Therefore, I believe a revised version of the fair use doctrine is necessary. For instance, what if the law only allowed companies to use public materials that were published more than two years ago?
---
5. How might this lawsuit between The New York Times and OpenAI impact the future development and regulation of AI technologies in various industries, considering their use of copyrighted material?
→ Policymakers should revise the current copyright laws, as they are not suitable for new technologies like AI unless updated.
---
#
Some food for thoughts:
1. Case Tracker: Artificial Intelligence, Copyrights and Class Actions (New York Times v. Microsoft) (https://www.bakerlaw.com/new-york-times-v-microsoft/)
2. "Nightshade", defensive tool for artists against AI art generators, poisons AI training models that scrape their work without permission (https://amt-lab.org/reviews/2023/11/nightshade-a-defensive-tool-for-artists-against-ai-art-generators)
---
IT기업 관련 커리어 조언 추천추천
https://devkingdom.tistory.com/383
기술블로그 목록
https://awesome-devblog.netlify.app/
BERT 관련 학습자료
https://jalammar.github.io/illustrated-transformer/
https://mchromiak.github.io/articles/2017/Sep/12/Transformer-Attention-is-all-you-need
#
https://hyeonseok.com/blog/900
REST Docs & Swagger UI
https://bbogle2.tistory.com/63
Google Cloud 에서 제공하는 컨테이너 기반의 Serverless 솔루션인 Cloud Run
https://tunz.kr/post/4 Transformer
https://mygumi.tistory.com/437
https://coding-factory.tistory.com/445
#커피컵홀더 #컵홀더 #슬리브 #sleeve #커피 #카페 #Cafe #coffee #드림라이크 #시청역커피 #광화문역커피 #을지로입구역커피 #작은카페 #꿈같은커피
#
1. Promise 함수 정리
https://mugglim.tistory.com/40
2. Servlet 기반으로 구동되는 웹서비스는 인증, 인가, 변환, 압축, 암호화 등 비즈니스 로직이 수행되기 전에 공통으로 사용되는 로직을 Servlet Filter로 구현한다. Servlet이 웹 요청을 처리하기 앞서 여러 Filter가 순서에 맞춰 수행되는 것이 Servlet 서버의 기본 동작 흐름
https://traeper.tistory.com/224
3. Java Object 클래스의 메소드 중에는 finalize() 라는 녀석이 있습니다. 아주 오래전 Java 태동기부터 존재했고 그 설계 의도도 명확했죠. 하지만 최근 몇년간 많은 문제점과 논란이 제기된 끝에 JDK 9 버전에서 Deprecated 되었고 결국 사라질 준비를 하고 있습니다. (JEP-421)
클래스의 생성(초기화) 시점에 리소스를 획득하고 종료 시점에 리소스를 다시 반납하는 패턴을 RAII(Resource Acqusition Is Initialization) 패턴 이라고 부릅니다. C++ 에서 주로 사용되는 패턴인데, GC 가 따로 없다보니 개발자가 직접 리소스를 할당하고 해제해주어야 했고 까딱 잘못하면 메모리 누수가 나곤했죠. 그래서 위 처럼 패턴을 만들어서 고착화 시킨겁니다.
https://jaeyeong951.medium.com/finalize-%EC%9D%80%ED%87%B4%EC%8B%9D-4a52fb855910
4. 검색 최적화
5. 유용한 사이트