Continuous Learning_Startup & Investment
https://www.whoop.com/ Whoop is a wearable fitness and recovery tracker that aims to provide insights into various aspects of your health and performance. It primarily focuses on tracking and analyzing metrics related to recovery, sleep, and strain. Hereโฆ
Anyone know how it compares to fitbit, garmin, oura ring and whoop?!
About quality of sleep.
Several factors can significantly impact the quality of sleep. Here are some key factors that can affect sleep quality:
1. Sleep Environment: Creating a comfortable sleep environment is crucial. Factors such as ambient temperature, noise levels, light exposure, and mattress and pillow quality can all influence sleep quality.
2. Sleep Duration: The amount of time you spend sleeping is essential. Consistently getting enough sleep that aligns with your individual needs is vital for optimal sleep quality.
3. Sleep Schedule and Consistency: Maintaining a regular sleep schedule, including consistent bedtimes and wake-up times, helps regulate the body's internal clock and promotes better sleep quality.
4. Sleep Hygiene: Practicing good sleep hygiene involves adopting habits that promote quality sleep. This includes having a relaxing pre-sleep routine, avoiding stimulants like caffeine and electronic devices close to bedtime, and creating a calm and peaceful sleep environment.
5. Stress and Mental Well-being: High levels of stress, anxiety, and certain mental health conditions can significantly impact sleep quality. Managing stress through relaxation techniques, mindfulness, and seeking support when needed can help improve sleep.
6. Physical Activity: Regular physical activity can promote better sleep quality. Engaging in moderate exercise during the day can help improve sleep onset and duration.
7. Diet and Nutrition: What you eat and drink can affect your sleep. Consuming heavy meals, stimulating substances like caffeine or nicotine, and excessive alcohol intake close to bedtime can disrupt sleep patterns.
8. Sleep Disorders and Medical Conditions: Certain sleep disorders, such as sleep apnea, insomnia, or restless legs syndrome, as well as underlying medical conditions, can interfere with sleep quality. Seeking appropriate medical guidance and treatment when necessary is important.
Several factors can significantly impact the quality of sleep. Here are some key factors that can affect sleep quality:
1. Sleep Environment: Creating a comfortable sleep environment is crucial. Factors such as ambient temperature, noise levels, light exposure, and mattress and pillow quality can all influence sleep quality.
2. Sleep Duration: The amount of time you spend sleeping is essential. Consistently getting enough sleep that aligns with your individual needs is vital for optimal sleep quality.
3. Sleep Schedule and Consistency: Maintaining a regular sleep schedule, including consistent bedtimes and wake-up times, helps regulate the body's internal clock and promotes better sleep quality.
4. Sleep Hygiene: Practicing good sleep hygiene involves adopting habits that promote quality sleep. This includes having a relaxing pre-sleep routine, avoiding stimulants like caffeine and electronic devices close to bedtime, and creating a calm and peaceful sleep environment.
5. Stress and Mental Well-being: High levels of stress, anxiety, and certain mental health conditions can significantly impact sleep quality. Managing stress through relaxation techniques, mindfulness, and seeking support when needed can help improve sleep.
6. Physical Activity: Regular physical activity can promote better sleep quality. Engaging in moderate exercise during the day can help improve sleep onset and duration.
7. Diet and Nutrition: What you eat and drink can affect your sleep. Consuming heavy meals, stimulating substances like caffeine or nicotine, and excessive alcohol intake close to bedtime can disrupt sleep patterns.
8. Sleep Disorders and Medical Conditions: Certain sleep disorders, such as sleep apnea, insomnia, or restless legs syndrome, as well as underlying medical conditions, can interfere with sleep quality. Seeking appropriate medical guidance and treatment when necessary is important.
Continuous Learning_Startup & Investment
About quality of sleep. Several factors can significantly impact the quality of sleep. Here are some key factors that can affect sleep quality: 1. Sleep Environment: Creating a comfortable sleep environment is crucial. Factors such as ambient temperatureโฆ
The ideal sleep environment can vary from person to person based on individual preferences and needs. However, here are some general guidelines:
Temperature: The ideal temperature for sleep tends to be cooler, typically between 60 to 67 degrees Fahrenheit (15 to 19 degrees Celsius). Experiment with different temperatures to find what feels most comfortable for you.
Noise Level: A quiet sleep environment is generally preferred. However, some individuals may find white noise or soothing sounds beneficial for relaxation. Use earplugs or a white noise machine if needed to create a peaceful environment.
Light Exposure: A dark environment promotes better sleep. Use blackout curtains or an eye mask to block out external light sources. Minimize exposure to electronic screens (e.g., smartphones, tablets) before bed as the blue light emitted can interfere with sleep.
Mattress and Pillow: Choose a mattress and pillow that provide adequate support and comfort for your body. The ideal firmness and type can vary based on personal preferences and any specific physical considerations you may have.
Sleep Duration: The ideal sleep duration varies across age groups. Adults generally require 7 to 9 hours of sleep per night, although individual variations exist. It's important to prioritize consistent and sufficient sleep to meet your individual needs.
Sleep Schedule: Establishing a consistent sleep schedule is beneficial. Aim to go to bed and wake up at the same time every day, including weekends, to regulate your body's internal clock. Consistency helps maintain a regular sleep-wake cycle.
Sleep Hygiene: Good sleep hygiene practices include establishing a relaxing bedtime routine, maintaining a comfortable sleep environment, avoiding stimulants (caffeine, nicotine) close to bedtime, and engaging in relaxing activities before sleep (e.g., reading, taking a warm bath).
Remember, individual preferences and needs may vary. It's important to listen to your body and adjust your sleep environment, schedule, and hygiene practices based on what helps you feel most rested and refreshed upon waking. If you have specific concerns or difficulties with sleep, consult with a healthcare professional for personalized guidance.
Temperature: The ideal temperature for sleep tends to be cooler, typically between 60 to 67 degrees Fahrenheit (15 to 19 degrees Celsius). Experiment with different temperatures to find what feels most comfortable for you.
Noise Level: A quiet sleep environment is generally preferred. However, some individuals may find white noise or soothing sounds beneficial for relaxation. Use earplugs or a white noise machine if needed to create a peaceful environment.
Light Exposure: A dark environment promotes better sleep. Use blackout curtains or an eye mask to block out external light sources. Minimize exposure to electronic screens (e.g., smartphones, tablets) before bed as the blue light emitted can interfere with sleep.
Mattress and Pillow: Choose a mattress and pillow that provide adequate support and comfort for your body. The ideal firmness and type can vary based on personal preferences and any specific physical considerations you may have.
Sleep Duration: The ideal sleep duration varies across age groups. Adults generally require 7 to 9 hours of sleep per night, although individual variations exist. It's important to prioritize consistent and sufficient sleep to meet your individual needs.
Sleep Schedule: Establishing a consistent sleep schedule is beneficial. Aim to go to bed and wake up at the same time every day, including weekends, to regulate your body's internal clock. Consistency helps maintain a regular sleep-wake cycle.
Sleep Hygiene: Good sleep hygiene practices include establishing a relaxing bedtime routine, maintaining a comfortable sleep environment, avoiding stimulants (caffeine, nicotine) close to bedtime, and engaging in relaxing activities before sleep (e.g., reading, taking a warm bath).
Remember, individual preferences and needs may vary. It's important to listen to your body and adjust your sleep environment, schedule, and hygiene practices based on what helps you feel most rested and refreshed upon waking. If you have specific concerns or difficulties with sleep, consult with a healthcare professional for personalized guidance.
โค1
Interesting github repo lists related to AI
https://github.com/ChristosChristofidis/awesome-deep-learning#courses
https://github.com/f/awesome-chatgpt-prompts
https://github.com/openai/gpt-3
https://github.com/lencx/ChatGPT
https://github.com/ml-tooling/best-of-ml-python
https://github.com/siyuan-note/siyuan
https://github.com/EleutherAI/gpt-neo
https://github.com/acheong08/EdgeGPT
https://github.com/BlinkDL/RWKV-LM
https://github.com/yetone/bob-plugin-openai-translator
https://github.com/arc53/DocsGPT
https://github.com/databrickslabs/dolly
https://github.com/waylaidwanderer/node-chatgpt-api
https://github.com/NVIDIA/FasterTransformer
https://github.com/stars/acheong08/lists/awesome-chatgpt
https://kh-kim.github.io/nlp_with_deep_learning_blog/docs/gradient-descent
http://introtodeeplearning.com/
https://flux.paradigm.xyz/
https://www.youtube.com/watch?v=sUn26PzgbJA
https://blog.naver.com/PostView.naver?blogId=tosoha1&logNo=223062904451&redirect=Dlog&widgetTypeCall=true&topReferer=https%3A%2F%2Fwww.naver.com%2Fmy.html&directAccess=false
https://arxiv.org/pdf/2303.12712.pdf?fbclid=IwAR1rLoqIrdUfKvoGIe65-asaPPjwLDsHJAPWE2mIW_yItyytECLmGgvQQFY
https://m.blog.naver.com/PostView.naver?blogId=tmdejr1267&logNo=223061304796&navType=by
https://vectara.com/
https://python.langchain.com/en/latest/
https://huggingface.co/spaces/JavaFXpert/Chat-GPT-LangChain/blob/main/app.py
https://www.mlq.ai/gpt-4-pinecone-website-ai-assistant/
https://theresanaiforthat.com/ai/literally-anything/
https://github.com/kuleshov/minillm
https://github.com/getcursor/cursor
https://www.youtube.com/watch?v=Ix9WIZpArm0
https://github.com/mayooear/gpt4-pdf-chatbot-langchain
https://www.youtube.com/@chatwithdata
https://arxiv.org/abs/2303.17564?fbclid=IwAR24SXQqKytV9BHnXqzXe6aDfaR4ZAl4Y4ViO7S8n-KUCqbtxsXE9Ayqp5g
https://arxiv.org/abs/2303.01037
https://medium.com/@yangyou_berkeley/colossalchat-an-open-source-solution-for-cloning-chatgpt-with-a-complete-rlhf-pipeline-5edf08fb538b
https://github.com/ChristosChristofidis/awesome-deep-learning#courses
https://github.com/f/awesome-chatgpt-prompts
https://github.com/openai/gpt-3
https://github.com/lencx/ChatGPT
https://github.com/ml-tooling/best-of-ml-python
https://github.com/siyuan-note/siyuan
https://github.com/EleutherAI/gpt-neo
https://github.com/acheong08/EdgeGPT
https://github.com/BlinkDL/RWKV-LM
https://github.com/yetone/bob-plugin-openai-translator
https://github.com/arc53/DocsGPT
https://github.com/databrickslabs/dolly
https://github.com/waylaidwanderer/node-chatgpt-api
https://github.com/NVIDIA/FasterTransformer
https://github.com/stars/acheong08/lists/awesome-chatgpt
https://kh-kim.github.io/nlp_with_deep_learning_blog/docs/gradient-descent
http://introtodeeplearning.com/
https://flux.paradigm.xyz/
https://www.youtube.com/watch?v=sUn26PzgbJA
https://blog.naver.com/PostView.naver?blogId=tosoha1&logNo=223062904451&redirect=Dlog&widgetTypeCall=true&topReferer=https%3A%2F%2Fwww.naver.com%2Fmy.html&directAccess=false
https://arxiv.org/pdf/2303.12712.pdf?fbclid=IwAR1rLoqIrdUfKvoGIe65-asaPPjwLDsHJAPWE2mIW_yItyytECLmGgvQQFY
https://m.blog.naver.com/PostView.naver?blogId=tmdejr1267&logNo=223061304796&navType=by
https://vectara.com/
https://python.langchain.com/en/latest/
https://huggingface.co/spaces/JavaFXpert/Chat-GPT-LangChain/blob/main/app.py
https://www.mlq.ai/gpt-4-pinecone-website-ai-assistant/
https://theresanaiforthat.com/ai/literally-anything/
https://github.com/kuleshov/minillm
https://github.com/getcursor/cursor
https://www.youtube.com/watch?v=Ix9WIZpArm0
https://github.com/mayooear/gpt4-pdf-chatbot-langchain
https://www.youtube.com/@chatwithdata
https://arxiv.org/abs/2303.17564?fbclid=IwAR24SXQqKytV9BHnXqzXe6aDfaR4ZAl4Y4ViO7S8n-KUCqbtxsXE9Ayqp5g
https://arxiv.org/abs/2303.01037
https://medium.com/@yangyou_berkeley/colossalchat-an-open-source-solution-for-cloning-chatgpt-with-a-complete-rlhf-pipeline-5edf08fb538b
GitHub
GitHub - ChristosChristofidis/awesome-deep-learning: A curated list of awesome Deep Learning tutorials, projects and communities.
A curated list of awesome Deep Learning tutorials, projects and communities. - ChristosChristofidis/awesome-deep-learning
https://www.instagram.com/p/CpcLIYpPsAY/?igshid=MjkzY2Y1YTY%3D
Hospitals face challenges when negotiating with insurance companies. Customers' problems present opportunities for businesses.
Both hospitals and individuals, including myself, face challenges in collecting necessary documents, applying for insurance, and negotiating with insurance companies.
Could we address this problem?
Hospitals face challenges when negotiating with insurance companies. Customers' problems present opportunities for businesses.
Both hospitals and individuals, including myself, face challenges in collecting necessary documents, applying for insurance, and negotiating with insurance companies.
Could we address this problem?
Google is investing in Runway, a New Yorkโbased startup that lets customers generate video from text descriptions using artificial intelligence it pioneered, at a valuation of around $1.5 billion including the new capital, according to two people familiar with the matter.
The investment, part of a roughly $100 million funding round, underscores the fierce competition among cloud providers to get close to companies with cutting-edge AI services that could become major cloud customers or acquisition targets in the future. Amazon Web Services has touted Runway, which generates relatively little revenue from its video-editing tools, as a key AI-startup customer but Runway is now expected to rent cloud servers from Google, said one person briefed about the deal.
https://www.theinformation.com/articles/google-invests-in-ai-startup-runway-to-wrest-cloud-business-from-aws?utm_campaign=article_email&utm_content=article-10624&utm_source=sg&utm_medium=email&rc=ocojsj
The investment, part of a roughly $100 million funding round, underscores the fierce competition among cloud providers to get close to companies with cutting-edge AI services that could become major cloud customers or acquisition targets in the future. Amazon Web Services has touted Runway, which generates relatively little revenue from its video-editing tools, as a key AI-startup customer but Runway is now expected to rent cloud servers from Google, said one person briefed about the deal.
https://www.theinformation.com/articles/google-invests-in-ai-startup-runway-to-wrest-cloud-business-from-aws?utm_campaign=article_email&utm_content=article-10624&utm_source=sg&utm_medium=email&rc=ocojsj
The Information
Google Invests in AI Startup Runway to Wrest Cloud Business From AWS
Google is investing in Runway, a New Yorkโbased startup that lets customers generate video from text descriptions using artificial intelligence it pioneered, at a valuation of around $1.5 billion including the new capital, according to two people familiarโฆ
AI ๋ํญํด์๋๋ค์ ใ
ใ
์ค๋ ๋ค์ ๐
๐1
ํ์ฅํ, ์ ์์ ํ, ์๋์ฐจ, ํจ์
&์ฌ์ , ์ํ, B2B ๋ถ์ผ์์ ํด์ธ ๋ง์ผํ
(์๋ฏธ๊ถ)์ ๋ด๋นํ์๋ ๋ถ ํน์ ์ง์ธ์ด ์์ผ์ค๊น์?
๊ด์ฌ์์ผ์ค๋งํ ์ฃผ์ ๋ฅผ ์ ์๋๋ฆฌ๋ ค๊ณ ํฉ๋๋ค!
@MatthewMinseokKim ์ผ๋ก ์ฐ๋ฝ์ฃผ์๋ฉด ๊ฐ์ฌํ๊ฒ ์ต๋๋ค ๐
๊ด์ฌ์์ผ์ค๋งํ ์ฃผ์ ๋ฅผ ์ ์๋๋ฆฌ๋ ค๊ณ ํฉ๋๋ค!
@MatthewMinseokKim ์ผ๋ก ์ฐ๋ฝ์ฃผ์๋ฉด ๊ฐ์ฌํ๊ฒ ์ต๋๋ค ๐
Continuous Learning_Startup & Investment
ํ์ฅํ, ์ ์์ ํ, ์๋์ฐจ, ํจ์
&์ฌ์ , ์ํ, B2B ๋ถ์ผ์์ ํด์ธ ๋ง์ผํ
(์๋ฏธ๊ถ)์ ๋ด๋นํ์๋ ๋ถ ํน์ ์ง์ธ์ด ์์ผ์ค๊น์? ๊ด์ฌ์์ผ์ค๋งํ ์ฃผ์ ๋ฅผ ์ ์๋๋ฆฌ๋ ค๊ณ ํฉ๋๋ค! @MatthewMinseokKim ์ผ๋ก ์ฐ๋ฝ์ฃผ์๋ฉด ๊ฐ์ฌํ๊ฒ ์ต๋๋ค ๐
๐ ์ฑ๋ด ์์ง ์ต์ ํ(Chatbot Engine Optimization)์ ์๋๊ฐ ์์ต๋๋ค. ์ค๋น๋์
จ๋์? ๐
ChatGPT๋ ํ์ฌ ํ๋ฃจ ํ์ฑ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ด 6000๋ง, ์๊ฐ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ ๋ฌด๋ ค 18์ต์ ์ด๋ฆ ๋๋ค. Yahoo, Google, Facebook, YouTube, TikTok ๋ฑ์ ์ฌ๋ก๋ฅผ ๋ดค์ ๋, ์ฌ์ฉ์๋ค์ด ๋ง์ ์๊ฐ์ ๋ณด๋ด๋ ์๋น์ค๋ ์๋ก์ด ๊ด๊ณ ํ๋ซํผ์ผ๋ก ์ฑ์ฅํด์์ต๋๋ค. ๋ง์ฐฌ๊ฐ์ง๋ก ์์ผ๋ก ๊ฐ์ฅ ํฐ ๊ด๊ณ ํ๋ซํผ์ ChatGPT์ ๊ฐ์ AI ์ฑ๋ด๋ค์ด ๋ ๊ฒ์ ๋๋ค.
๊ณผ๊ฑฐ์ Facebook, Youtube, TikTok์ด ์ด๊ธฐ์ ๋ฑ์ฅํ์ ๋ ๋ง์ผํ ์ ํจ๊ณผ์ ์ผ๋ก ์งํํ ๊ธฐ์ ๋ค์ด ์๋ก์ด ์ฑ์ฅ์ ๊ธฐํ๋ฅผ ์ก์๋ ๊ฒ์ฒ๋ผ, AI Chatbot์ ํตํด ์์ ์ ์ ํ ์ ๋ณด๋ฅผ ํจ๊ณผ์ ์ผ๋ก ์ ๋ฌํ๋ ๊ฒ์ ๋ชจ๋ ๊ธฐ์ ์๊ฒ ์ค์ํ ๊ธฐํ๊ฐ ๋ ์ ์์ต๋๋ค.
์ด๋ฏธ Bay Area์ ๋ฒ๋ฅ ์ฌ๋ฌด์์ ์ ์์๊ฑฐ๋ ํ์ฌ๋ค ์ค ChatGPT๋ฅผ ํตํด ์ ๊ท ๊ณ ๊ฐ์ ํ๋ณดํ ๊ฒฝ์ฐ๊ฐ ์์ต๋๋ค. AI ์ฑ๋ด ์์ฅ์์ด๋ฏธ ๋น ๋ฅด๊ฒ ์ฑ์ฅํ๊ณ ์์ผ๋ฉฐ, ์ด๋ ๋์งํธ ๋ง์ผํ ์์ SEO๋งํผ ์ค์ํ ์ญํ ์ ํ๊ฒ ๋ ๊ฒ์ ๋๋ค.
SEO๊ฐ ์ด๋ฏธ ํฌํ๋ ๊ฒฝ์ ์์ฅ์ธ ๋ฐ๋ฉด, ์ฑ๋ด ์์ง ์ต์ ํ(ChatEO)์ ์๋๋ ๋ง๋ฐ์ง๋ฅผ ์์ํ์ต๋๋ค. ChatEO๋ AI๊ฐ ํน์ ์ ํ์ ์ ๋ณด๋ฅผ ์ ํ์ตํ๊ณ ๊ฒ์ํ ์ ์๋๋ก ์ ๋ณด๋ฅผ ์ ์ ๋นํ๊ณ , ํน์ ์ ํ์ ์๋นํ๋ ๊ณ ๊ฐ์๊ฒ ๋ง๋ AI ๊ธฐ๋ฐ ์๋น์ค๋ฅผ ์ค๋นํ๋ ๊ฒ์ ์๋ฏธํฉ๋๋ค.
๋์งํธ ๋ง์ผํ ์ ์์ด์ ํฅ๋ฏธ์ง์งํ ์๊ธฐ์ ๋๋ค. AI ์ฑ๋ด๊ณผ ChatEO์ ๋ถ์์ ๊ณ ๊ฐ ์ฐธ์ฌ์ ๋์งํธ ๊ด๊ณ ์ ์งํ๋๋ฅผ ํ์ ์ ์ผ๋ก ๋ฐ๊ฟ ์ ์์ ๊ฒ์ ๋๋ค. ์ด ๋ณํ๋ฅผ ํจ๊ณผ์ ์ผ๋ก ๋ค๋ฃฐ ์ ์๋ ๊ธฐ์ ๋ค์ ์ด ์๋ก์ด ํ๊ฒฝ์์ ๋จ์ํ ์์กดํ๋ ๊ฒ์ ๋์ด ๋ฒ์ํ ์ ์์ ๊ฒ ์ ๋๋ค.
ํ๊ตญ์ด ๊ธ๋ก๋ฒํ๊ฒ ์์ถํ๋ ํ๋ชฉ์ธ ํ์ฅํ, ์ ์์ ํ, ์๋์ฐจ, ํจ์ &์ฌ์ , ์ํ, B2B ๋ฑ์ด๊ณ ๊ด๋ จ๋ ์ ๊ณ์์ ์๋ฏธ๊ถ ๋์ ํด์ธ ๋ง์ผํ ์ ๋ด๋นํ์๋ ๋ถ๋ค์ ์ด ํธ๋ ๋๋ฅผ ์ ํํ ์ดํดํ ํ์๊ฐ ์์ต๋๋ค.
์ด ์๋ก์ด ํธ๋ ๋์ ๊ด์ฌ์ด ์์ผ์ ๋ถ๋ค, ํน์ ๋ ์์ธํ ์ด์ผ๊ธฐ๋ฅผ ๋๋๊ณ ์ถ์ผ์ ๋ถ๋ค์ ์๋ ๋งํฌ๋ฅผ ํด๋ฆญํด์ฃผ์๊ธฐ ๋ฐ๋๋๋ค. ๐
https://forms.gle/vqChSP989extfTLj7
#AI #์ฑ๋ด #ChatGPT #ChatEO #๋์งํธ๋ง์ผํ #SEO
ChatGPT๋ ํ์ฌ ํ๋ฃจ ํ์ฑ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ด 6000๋ง, ์๊ฐ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ ๋ฌด๋ ค 18์ต์ ์ด๋ฆ ๋๋ค. Yahoo, Google, Facebook, YouTube, TikTok ๋ฑ์ ์ฌ๋ก๋ฅผ ๋ดค์ ๋, ์ฌ์ฉ์๋ค์ด ๋ง์ ์๊ฐ์ ๋ณด๋ด๋ ์๋น์ค๋ ์๋ก์ด ๊ด๊ณ ํ๋ซํผ์ผ๋ก ์ฑ์ฅํด์์ต๋๋ค. ๋ง์ฐฌ๊ฐ์ง๋ก ์์ผ๋ก ๊ฐ์ฅ ํฐ ๊ด๊ณ ํ๋ซํผ์ ChatGPT์ ๊ฐ์ AI ์ฑ๋ด๋ค์ด ๋ ๊ฒ์ ๋๋ค.
๊ณผ๊ฑฐ์ Facebook, Youtube, TikTok์ด ์ด๊ธฐ์ ๋ฑ์ฅํ์ ๋ ๋ง์ผํ ์ ํจ๊ณผ์ ์ผ๋ก ์งํํ ๊ธฐ์ ๋ค์ด ์๋ก์ด ์ฑ์ฅ์ ๊ธฐํ๋ฅผ ์ก์๋ ๊ฒ์ฒ๋ผ, AI Chatbot์ ํตํด ์์ ์ ์ ํ ์ ๋ณด๋ฅผ ํจ๊ณผ์ ์ผ๋ก ์ ๋ฌํ๋ ๊ฒ์ ๋ชจ๋ ๊ธฐ์ ์๊ฒ ์ค์ํ ๊ธฐํ๊ฐ ๋ ์ ์์ต๋๋ค.
์ด๋ฏธ Bay Area์ ๋ฒ๋ฅ ์ฌ๋ฌด์์ ์ ์์๊ฑฐ๋ ํ์ฌ๋ค ์ค ChatGPT๋ฅผ ํตํด ์ ๊ท ๊ณ ๊ฐ์ ํ๋ณดํ ๊ฒฝ์ฐ๊ฐ ์์ต๋๋ค. AI ์ฑ๋ด ์์ฅ์์ด๋ฏธ ๋น ๋ฅด๊ฒ ์ฑ์ฅํ๊ณ ์์ผ๋ฉฐ, ์ด๋ ๋์งํธ ๋ง์ผํ ์์ SEO๋งํผ ์ค์ํ ์ญํ ์ ํ๊ฒ ๋ ๊ฒ์ ๋๋ค.
SEO๊ฐ ์ด๋ฏธ ํฌํ๋ ๊ฒฝ์ ์์ฅ์ธ ๋ฐ๋ฉด, ์ฑ๋ด ์์ง ์ต์ ํ(ChatEO)์ ์๋๋ ๋ง๋ฐ์ง๋ฅผ ์์ํ์ต๋๋ค. ChatEO๋ AI๊ฐ ํน์ ์ ํ์ ์ ๋ณด๋ฅผ ์ ํ์ตํ๊ณ ๊ฒ์ํ ์ ์๋๋ก ์ ๋ณด๋ฅผ ์ ์ ๋นํ๊ณ , ํน์ ์ ํ์ ์๋นํ๋ ๊ณ ๊ฐ์๊ฒ ๋ง๋ AI ๊ธฐ๋ฐ ์๋น์ค๋ฅผ ์ค๋นํ๋ ๊ฒ์ ์๋ฏธํฉ๋๋ค.
๋์งํธ ๋ง์ผํ ์ ์์ด์ ํฅ๋ฏธ์ง์งํ ์๊ธฐ์ ๋๋ค. AI ์ฑ๋ด๊ณผ ChatEO์ ๋ถ์์ ๊ณ ๊ฐ ์ฐธ์ฌ์ ๋์งํธ ๊ด๊ณ ์ ์งํ๋๋ฅผ ํ์ ์ ์ผ๋ก ๋ฐ๊ฟ ์ ์์ ๊ฒ์ ๋๋ค. ์ด ๋ณํ๋ฅผ ํจ๊ณผ์ ์ผ๋ก ๋ค๋ฃฐ ์ ์๋ ๊ธฐ์ ๋ค์ ์ด ์๋ก์ด ํ๊ฒฝ์์ ๋จ์ํ ์์กดํ๋ ๊ฒ์ ๋์ด ๋ฒ์ํ ์ ์์ ๊ฒ ์ ๋๋ค.
ํ๊ตญ์ด ๊ธ๋ก๋ฒํ๊ฒ ์์ถํ๋ ํ๋ชฉ์ธ ํ์ฅํ, ์ ์์ ํ, ์๋์ฐจ, ํจ์ &์ฌ์ , ์ํ, B2B ๋ฑ์ด๊ณ ๊ด๋ จ๋ ์ ๊ณ์์ ์๋ฏธ๊ถ ๋์ ํด์ธ ๋ง์ผํ ์ ๋ด๋นํ์๋ ๋ถ๋ค์ ์ด ํธ๋ ๋๋ฅผ ์ ํํ ์ดํดํ ํ์๊ฐ ์์ต๋๋ค.
์ด ์๋ก์ด ํธ๋ ๋์ ๊ด์ฌ์ด ์์ผ์ ๋ถ๋ค, ํน์ ๋ ์์ธํ ์ด์ผ๊ธฐ๋ฅผ ๋๋๊ณ ์ถ์ผ์ ๋ถ๋ค์ ์๋ ๋งํฌ๋ฅผ ํด๋ฆญํด์ฃผ์๊ธฐ ๋ฐ๋๋๋ค. ๐
https://forms.gle/vqChSP989extfTLj7
#AI #์ฑ๋ด #ChatGPT #ChatEO #๋์งํธ๋ง์ผํ #SEO
Google Docs
ChatbotEO(Chatbot Engine Optimization) ๋ฌธ์
๐ ์ฑ๋ด ์์ง ์ต์ ํ(Chatbot Engine Optimization)์ ์๋๊ฐ ์์ต๋๋ค. ์ค๋น๋์
จ๋์? ๐
ChatGPT๋ ํ์ฌ ํ๋ฃจ ํ์ฑ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ด 6000๋ง, ์๊ฐ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ ๋ฌด๋ ค 18์ต์ ์ด๋ฆ ๋๋ค. Yahoo, Google, Facebook, YouTube, TikTok ๋ฑ์ ์ฌ๋ก๋ฅผ ๋ดค์ ๋, ์ฌ์ฉ์๋ค์ด ๋ง์ ์๊ฐ์ ๋ณด๋ด๋ ์๋น์ค๋ ์๋ก์ด ๊ด๊ณ ํ๋ซํผ์ผ๋ก ์ฑ์ฅํด์์ต๋๋ค. ๋ง์ฐฌ๊ฐ์ง๋ก ์์ผ๋ก ๊ฐ์ฅ ํฐ ๊ด๊ณ ํ๋ซํผ์ ChatGPT์ ๊ฐ์ AIโฆ
ChatGPT๋ ํ์ฌ ํ๋ฃจ ํ์ฑ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ด 6000๋ง, ์๊ฐ ์ฌ์ฉ์ ๋ฐฉ๋ฌธ์ ๋ฌด๋ ค 18์ต์ ์ด๋ฆ ๋๋ค. Yahoo, Google, Facebook, YouTube, TikTok ๋ฑ์ ์ฌ๋ก๋ฅผ ๋ดค์ ๋, ์ฌ์ฉ์๋ค์ด ๋ง์ ์๊ฐ์ ๋ณด๋ด๋ ์๋น์ค๋ ์๋ก์ด ๊ด๊ณ ํ๋ซํผ์ผ๋ก ์ฑ์ฅํด์์ต๋๋ค. ๋ง์ฐฌ๊ฐ์ง๋ก ์์ผ๋ก ๊ฐ์ฅ ํฐ ๊ด๊ณ ํ๋ซํผ์ ChatGPT์ ๊ฐ์ AIโฆ
Continuous Learning_Startup & Investment
ํ์ฅํ, ์ ์์ ํ, ์๋์ฐจ, ํจ์
&์ฌ์ , ์ํ, B2B ๋ถ์ผ์์ ํด์ธ ๋ง์ผํ
(์๋ฏธ๊ถ)์ ๋ด๋นํ์๋ ๋ถ ํน์ ์ง์ธ์ด ์์ผ์ค๊น์? ๊ด์ฌ์์ผ์ค๋งํ ์ฃผ์ ๋ฅผ ์ ์๋๋ฆฌ๋ ค๊ณ ํฉ๋๋ค! @MatthewMinseokKim ์ผ๋ก ์ฐ๋ฝ์ฃผ์๋ฉด ๊ฐ์ฌํ๊ฒ ์ต๋๋ค ๐
What Charlie said.
์๊ธฐ ์์ฃผ ํธํฅ์ ์ ์ํ๋ผ
๋ง์ง๋ง์ โ์๊ธฐ ์์ฃผ ํธํฅ(Self-serving bias)โ์ด๋ค. ์ฌ๋๋ค์ ์ ์ฒ์ ์ผ๋ก ๋ค๋ฅธ ์ฌ๋์ ์ด์ต๋ณด๋ค๋ ์๊ธฐ ์์ ์ ์ด์ต์๋ง ์ง์คํ๋ ๊ฒฝํฅ์ด ์๋ค. ์ฌ์ง์ด ์์ ์๊ฒ ์ข์ ์ผ์ด๋ฉด ์ฌํ์๋ ์ข๋ค๋ ์ฌ๊ณ ๋ฐฉ์์ด ๋ฌด์์์ ๊น๋ ค ์๊ธฐ ๋๋ฌธ์ ์ด๊ธฐ์ ์ธ ํ๋์ผ๋ก ์ด์ด์ง๊ธฐ๋ ํ๋ค.
ํ์ง๋ง, ๋ฉ๊ฑฐ๋ ์ด๊ฑด ๋๋จํ ์๋ชป๋ ์ฌ๊ณ ๋ฐฉ์์ด๋ผ๋ฉฐ ํ๋ช ํด์ง๊ธฐ ์ํด์๋ ์๊ธฐ ์์ฃผ ํธํฅ์์ ๋ฒ์ด๋์ผ ํ์ง๋ง, ๋๋ค์ ์ฌ๋๋ค์ด ์๊ธฐ ์์ฃผ ํธํฅ์ ๊ฐ์ง๊ณ ์๋ค๋ ์ฌ์ค๋ ์ธ์ ํด์ผ ํ๋ค๊ณ ๋งํ๋ค. ์๊ธฐ ์์ฃผ ํธํฅ์ ๋ฒ์ด๋๊ธฐ๋ ๋๋จํ ์ด๋ ต๊ธฐ ๋๋ฌธ์ด๋ค.
๊ทธ๋์ ๋ฉ๊ฑฐ๋ ์ฐ๋ฆฌ๊ฐ ํ๋ํ ๋๋ ์ฌ๋๋ค์ด ์๊ธฐ ์์ฃผ ํธํฅ์ ๊ฐ์ง๊ณ ์๋ค๋ ๊ฑธ ์ผ๋์ ๋๊ณ ํ๋ํด์ผ ํ๋ค๋ ์ ์ ์๋ฆฌํ๊ฒ ์ง์ ํ๋ค. ์ฌ๊ธฐ์ ๋ฉ๊ฑฐ๊ฐ ์๋ก ๋ ๊ฑด ๋ฏธ๊ตญ ํฌ์์ํ ์ด๋ก๋จผ ๋ธ๋ผ๋์ค์ ๊ณ ๋ฌธ๋ณํธ์ฌ๋ค.
๋น์ ์ด๋ก๋จผ ๋ธ๋ผ๋์ค์ ์ต๊ณ ๊ฒฝ์์(CEO)๊ฐ ์์ฌ ์ฑ๊ถ ํธ๋ ์ด๋๊ฐ ๊ตญ์ฑ ์ ์ฐฐ์์ ๊ด๋ จ ๊ท์ ์ ์ด๊ฒผ๋ค๋ ๊ฑธ ์๊ฒ ๋๋ค. ๊ณ ๋ฌธ๋ณํธ์ฌ๋ CEO์๊ฒ ๋น๊ตญ์ ๋ณด๊ณ ํ ๋ฒ์ ์๋ฌด๋ ์์ง๋ง, ๊ทธ๋๋ ๋ณด๊ณ ํ๋ ๊ฒ ๋๋์ ์ธ ์๋ฌด๋ผ๊ณ ๋งํ๋ค. ํ์ง๋ง, ์ด๊ฑด ๋๊ตฌ๋ ์์น ์๋ ์ผ์ด์๊ธฐ์ CEO๋ ๊ณ์ ๋ฏธ๋ค๊ณ ๊ฒฐ๊ตญ ์ด๋ก๋จผ ๋ธ๋ฌ๋์ค์ ์ฑ๊ถ ์ค์บ๋ค๋ก ๋นํ๋๋ฉด์ CEO๋ฟ ์๋๋ผ ๊ณ ๋ฌธ๋ณํธ์ฌ๊น์ง ์ผ์๋ฆฌ๋ฅผ ์์๋ค.
์ด์ ๋ํด, ๋ฉ๊ฑฐ๋ ๋ฏธ๊ตญ ๊ฑด๊ตญ์ ์๋ฒ์ง ๋ฒค์๋ฏผ ํ๋ญํด๋ฆฐ์ ๋ง์ ๋น๋ ค โ๋ง์ฝ ์ค๋ํ๋ ค๊ณ ํ๋ค๋ฉด ์ด์ฑ์ด ์๋๋ผ ์ด์ต์ ํธ์ํ๋ผโ๊ณ ๋งํ๋ค. ์ฆ, CEO์ ์๊ธฐ ์์ฃผ ํธํฅ์ ๊ณ ๋ คํด์ CEO์ ์ด์ต์ ํธ์ํด์ผ ํ๋ค๋ ์๋ฏธ๋ค.
๋ง์ฝ ๊ณ ๋ฌธ๋ณํธ์ฌ๊ฐ CEO์๊ฒ โ์ด ์ผ์ ๊ทธ๋๋ก ๋๋๋ฉด ๊ฒฐ๊ตญ ์ฌ์ฅ๋์ ๊ฒฝ๋ ฅ์ ๋ง๊ฐ๋จ๋ฆฌ๊ณ ๋๊ณผ ์ง์, ๋ช ์๋ฅผ ๋ฌด๋๋จ๋ฆด ์ ์์ต๋๋ค. ์ ๊ถ๊ณ ๋ฅผ ๋ฐ๋ฅด๋ฉด ๊ทธ๋ฐ ์ฌํ๋ฅผ ๋ฐฉ์งํ ์ ์์ ๊ฒ๋๋คโ๋ผ๊ณ ๋งํ๋ค๋ฉด CEO์ ๊ณ ๋ฌธ๋ณํธ์ฌ ๋ชจ๋ ์๋ฌด์ผ ์์์ ๊ฒ์ด๋ค.
์๊ธฐ ์์ฃผ ํธํฅ์ ์ ์ํ๋ผ
๋ง์ง๋ง์ โ์๊ธฐ ์์ฃผ ํธํฅ(Self-serving bias)โ์ด๋ค. ์ฌ๋๋ค์ ์ ์ฒ์ ์ผ๋ก ๋ค๋ฅธ ์ฌ๋์ ์ด์ต๋ณด๋ค๋ ์๊ธฐ ์์ ์ ์ด์ต์๋ง ์ง์คํ๋ ๊ฒฝํฅ์ด ์๋ค. ์ฌ์ง์ด ์์ ์๊ฒ ์ข์ ์ผ์ด๋ฉด ์ฌํ์๋ ์ข๋ค๋ ์ฌ๊ณ ๋ฐฉ์์ด ๋ฌด์์์ ๊น๋ ค ์๊ธฐ ๋๋ฌธ์ ์ด๊ธฐ์ ์ธ ํ๋์ผ๋ก ์ด์ด์ง๊ธฐ๋ ํ๋ค.
ํ์ง๋ง, ๋ฉ๊ฑฐ๋ ์ด๊ฑด ๋๋จํ ์๋ชป๋ ์ฌ๊ณ ๋ฐฉ์์ด๋ผ๋ฉฐ ํ๋ช ํด์ง๊ธฐ ์ํด์๋ ์๊ธฐ ์์ฃผ ํธํฅ์์ ๋ฒ์ด๋์ผ ํ์ง๋ง, ๋๋ค์ ์ฌ๋๋ค์ด ์๊ธฐ ์์ฃผ ํธํฅ์ ๊ฐ์ง๊ณ ์๋ค๋ ์ฌ์ค๋ ์ธ์ ํด์ผ ํ๋ค๊ณ ๋งํ๋ค. ์๊ธฐ ์์ฃผ ํธํฅ์ ๋ฒ์ด๋๊ธฐ๋ ๋๋จํ ์ด๋ ต๊ธฐ ๋๋ฌธ์ด๋ค.
๊ทธ๋์ ๋ฉ๊ฑฐ๋ ์ฐ๋ฆฌ๊ฐ ํ๋ํ ๋๋ ์ฌ๋๋ค์ด ์๊ธฐ ์์ฃผ ํธํฅ์ ๊ฐ์ง๊ณ ์๋ค๋ ๊ฑธ ์ผ๋์ ๋๊ณ ํ๋ํด์ผ ํ๋ค๋ ์ ์ ์๋ฆฌํ๊ฒ ์ง์ ํ๋ค. ์ฌ๊ธฐ์ ๋ฉ๊ฑฐ๊ฐ ์๋ก ๋ ๊ฑด ๋ฏธ๊ตญ ํฌ์์ํ ์ด๋ก๋จผ ๋ธ๋ผ๋์ค์ ๊ณ ๋ฌธ๋ณํธ์ฌ๋ค.
๋น์ ์ด๋ก๋จผ ๋ธ๋ผ๋์ค์ ์ต๊ณ ๊ฒฝ์์(CEO)๊ฐ ์์ฌ ์ฑ๊ถ ํธ๋ ์ด๋๊ฐ ๊ตญ์ฑ ์ ์ฐฐ์์ ๊ด๋ จ ๊ท์ ์ ์ด๊ฒผ๋ค๋ ๊ฑธ ์๊ฒ ๋๋ค. ๊ณ ๋ฌธ๋ณํธ์ฌ๋ CEO์๊ฒ ๋น๊ตญ์ ๋ณด๊ณ ํ ๋ฒ์ ์๋ฌด๋ ์์ง๋ง, ๊ทธ๋๋ ๋ณด๊ณ ํ๋ ๊ฒ ๋๋์ ์ธ ์๋ฌด๋ผ๊ณ ๋งํ๋ค. ํ์ง๋ง, ์ด๊ฑด ๋๊ตฌ๋ ์์น ์๋ ์ผ์ด์๊ธฐ์ CEO๋ ๊ณ์ ๋ฏธ๋ค๊ณ ๊ฒฐ๊ตญ ์ด๋ก๋จผ ๋ธ๋ฌ๋์ค์ ์ฑ๊ถ ์ค์บ๋ค๋ก ๋นํ๋๋ฉด์ CEO๋ฟ ์๋๋ผ ๊ณ ๋ฌธ๋ณํธ์ฌ๊น์ง ์ผ์๋ฆฌ๋ฅผ ์์๋ค.
์ด์ ๋ํด, ๋ฉ๊ฑฐ๋ ๋ฏธ๊ตญ ๊ฑด๊ตญ์ ์๋ฒ์ง ๋ฒค์๋ฏผ ํ๋ญํด๋ฆฐ์ ๋ง์ ๋น๋ ค โ๋ง์ฝ ์ค๋ํ๋ ค๊ณ ํ๋ค๋ฉด ์ด์ฑ์ด ์๋๋ผ ์ด์ต์ ํธ์ํ๋ผโ๊ณ ๋งํ๋ค. ์ฆ, CEO์ ์๊ธฐ ์์ฃผ ํธํฅ์ ๊ณ ๋ คํด์ CEO์ ์ด์ต์ ํธ์ํด์ผ ํ๋ค๋ ์๋ฏธ๋ค.
๋ง์ฝ ๊ณ ๋ฌธ๋ณํธ์ฌ๊ฐ CEO์๊ฒ โ์ด ์ผ์ ๊ทธ๋๋ก ๋๋๋ฉด ๊ฒฐ๊ตญ ์ฌ์ฅ๋์ ๊ฒฝ๋ ฅ์ ๋ง๊ฐ๋จ๋ฆฌ๊ณ ๋๊ณผ ์ง์, ๋ช ์๋ฅผ ๋ฌด๋๋จ๋ฆด ์ ์์ต๋๋ค. ์ ๊ถ๊ณ ๋ฅผ ๋ฐ๋ฅด๋ฉด ๊ทธ๋ฐ ์ฌํ๋ฅผ ๋ฐฉ์งํ ์ ์์ ๊ฒ๋๋คโ๋ผ๊ณ ๋งํ๋ค๋ฉด CEO์ ๊ณ ๋ฌธ๋ณํธ์ฌ ๋ชจ๋ ์๋ฌด์ผ ์์์ ๊ฒ์ด๋ค.
๐3
New world to VR?
Appleโs first foray into mixed reality hardware (that is, a blend of real and virtual worlds) is one of the companyโs worst-kept secrets. The tech giant has been acquiring headset-friendly startups for years, and Tim Cook hasnโt been shy about his interests in augmented and virtual reality. Now, though, a headset appears to be ready for a WWDC debut โ Appleโs event logo even hints at a lens.
If the rumors are accurate, Appleโs stand-alone device (possibly called Reality Pro) may be more powerful than the Meta Quest Pro and many other high-end headsets. It might offer a 4K resolution per eye, with at least six cameras providing a view of the outside world as well as full body motion tracking. It could also use an M2 chip that outperforms the mobile processors you typically see in stand-alone headsets. Expectations are that it will require an external battery pack and last for just two hours on a charge, but it would be relatively light and slim.
The true star might be the software, though. The Apple headset will reportedly run a new platform (possibly called xrOS) designed with mixed reality in mind. Rather than rely on controllers, you would use hand gestures, gaze detection and Siri voice commands to navigate the interface. An Apple Watch-style crown would toggle between AR and VR, and you could type in mid-air.
Any news about AI from Apple? ๐ง
Appleโs first foray into mixed reality hardware (that is, a blend of real and virtual worlds) is one of the companyโs worst-kept secrets. The tech giant has been acquiring headset-friendly startups for years, and Tim Cook hasnโt been shy about his interests in augmented and virtual reality. Now, though, a headset appears to be ready for a WWDC debut โ Appleโs event logo even hints at a lens.
If the rumors are accurate, Appleโs stand-alone device (possibly called Reality Pro) may be more powerful than the Meta Quest Pro and many other high-end headsets. It might offer a 4K resolution per eye, with at least six cameras providing a view of the outside world as well as full body motion tracking. It could also use an M2 chip that outperforms the mobile processors you typically see in stand-alone headsets. Expectations are that it will require an external battery pack and last for just two hours on a charge, but it would be relatively light and slim.
The true star might be the software, though. The Apple headset will reportedly run a new platform (possibly called xrOS) designed with mixed reality in mind. Rather than rely on controllers, you would use hand gestures, gaze detection and Siri voice commands to navigate the interface. An Apple Watch-style crown would toggle between AR and VR, and you could type in mid-air.
Any news about AI from Apple? ๐ง
ํ๋ก์ธ์ค ๊ฐ๋
์ ํตํ ์ํ์ ์ถ๋ก ํฅ์
https://openai.com/research/improving-mathematical-reasoning-with-process-supervision?fbclid=IwAR3ICUhp_ALBMBUMOG_vqe1cqCMpLNTzm0yWkbIP2hbeeKX7OF-11_RCXQo
OpenAI๊ฐ ๋ ๋ค๋ฅธ ๋ํ๊ตฌ๋ฅผ ๋ง๋ จํ๊ตฐ์! ์ต์ข ์ ๋ต๋ฟ๋ง ์๋๋ผ ๊ฐ ์ ๋ต ๋จ๊ณ๋ง๋ค ๋ณด์์ ์ ๊ณตํ๋ ์ ๊ทผ์ผ๋ก ์ํ ๋ฌธ์ ํ์ด์์ ์๋ก์ด ๊ธฐ๋ก์ ์ธ์ด ๋ชจ๋ธ์ ํ๋ จ์์ผฐ ๋ค๊ณ ํฉ๋๋ค. ์ฐ๋ฆฌ๊ฐ ํ๊ต์์ ์ํ์ ๋ฐฐ์ฐ๋ ๋ฐฉ์๊ณผ ๋น์ทํ์ฃ ? ์ต์ข ์ ๋ต์ ๋งํ AI์ ๋ฑ์ ๋๋๋ฆฌ๋ ๋์ , ์ด์ ์ฌ๋ฐ๋ฅธ ๋จ๊ณ๋ฅผ ๋ฐ์ ๋๋ง๋ค ๋ณด์์ ์ ๊ณตํฉ๋๋ค. ๋ง์น ์ํ ์ ์๋์ด ์ต์ข ์ ๋ต๋ฟ๋ง ์๋๋ผ ํ์ต ๊ณผ์ ๊น์ง ์ ๊ฒฝ ์จ์ฃผ๋ ๊ฒ์ฒ๋ผ์. ์ด๋ฅผ '๊ณผ์ ๊ฐ๋ (Process Supervision)'์ด๋ผ๊ณ ๋ถ๋ฅด๋๋ฐ, ์ด๋ ํ๊ธฐ์ ์ธ ๋ณํ์ ๋๋ค. ๐งฎ๐ก
* ๋ ผ๋ฌธ์ ์ ๋ชฉ์ "Letโs Verify Step by Step"์ ๋๋ค. (์๋ ์ CoT - '์๊ฐ์ ์ฌ์ฌ' ๋ถ์ ์ผ์ผ์ผฐ๋ ๋ง๋ฒ์ ํํ Step by Step ์ ๋๋ค)
https://cdn.openai.com/improving-mathematical-reasoning-with-process-supervision/Lets_Verify_Step_by_Step.pdf?fbclid=IwAR1LojRtgWmMYERWatOlHsD_9AAxalBPDd2i9WiDfRwV_F_Scm3eT2IUB1k
์ด๊ฒ์ ๋จ์ํ ์ํ ๋ฌธ์ ๋ฅผ ๋ ์ ํธ๋ ๊ฒ๋ง์ด ์๋๋๋ค. ํ๋ก์ธ์ค ๊ฐ๋ ์ ๋ชจ๋ธ์ด ์ฌ๋์ด ์น์ธํ ํ๋ก์ธ์ค๋ฅผ ๋ฐ๋ผ ์ฐ๋ฆฌ์ ๋ ๋น์ทํ๊ฒ ์๊ฐํ๋๋ก ๋์์ค๋๋ค. ๋ง์น ์ธ๊ณต์ง๋ฅ์ ๋ง์์ ๋ค์ฌ๋ค๋ณด๊ณ ์๋ํ๋ ๊ณผ์ ์ ์ง์ผ๋ณด๋ ๊ฒ๊ณผ ๋น์ทํฉ๋๋ค. ์ด ๋ฐฉ๋ฒ์ ๋งค์ฐ ์ ๋ขฐํ ์ ์๋ ๊ฒ์ผ๋ก ์ ์ฆ๋์์ผ๋ฉฐ, AI๋ฅผ ๋ ์ ์ดํดํ๊ณ ์ฐ๋ฆฌ์ ๋ชฉํ์ ๋ถํฉํ๋๋ก ๋ง๋ค ์๋ ์์ต๋๋ค.
์ด๊ฒ์ AI์ ์๋๋ฅผ ๋ฆ์ถ๋ ๊ฒ์ด ์๋๋ผ ์คํ๋ ค ๋ ๋์ ์ฑ๋ฅ์ ์ ๊ณตํฉ๋๋ค! ์์ ๊ณผ ์ฑ๋ฅ ๋ชจ๋์์ ์์์ธ ์ ์ด์ฃ .
๋งค์ฐ ๊น๋ค๋ก์ด ์ํ ๋ฌธ์ ๋ฅผ ํ ์คํธํ ๊ฒฐ๊ณผ, '๊ณผ์ ๊ฐ๋ ํ' AI๊ฐ '๊ฒฐ๊ณผ ๊ฐ๋ ํ' AI๋ณด๋ค ์ ๋ฐ์ ์ผ๋ก ๋ ๋ฐ์ด๋ ์ฑ๋ฅ์ ๋ณด์์ต๋๋ค. ์ฌ์ง์ด ๋๋ถ๋ถ์ ์๋๊ฐ ์คํจ๋ก ๋์๊ฐ ์ด๋ ค์ด ์ผ๊ฐํ ๋ฌธ์ ์ ๋ํ ์ ํจํ ํด๊ฒฐ์ฑ ์ ์ฐพ์๋ด๊ธฐ๋ ํ์ต๋๋ค! ๐ฎ
ํ์ง๋ง ์ด ์ฑ๊ณผ๋ฅผ ์นญ์ฐฌํ๋๋ฐ๋ง ๊ทธ์น๋ ๊ฒ์ด ์๋๋ผ ์ง์งํ๊ฒ ๊ณ ๋ฏผํด์ผ ํ ๋ถ๋ถ์ด ์์ต๋๋ค. ๐ค
์ด ๋ฐฉ๋ฒ์ ๋ค๋ฅธ ๋ฌธ์ ํด๊ฒฐ ์์ ์์ AI๋ฅผ ํ๋ จํ๋ ๋ฐ ์ฌ์ฉํ ์ ์์๊น์? ๊ทธ๋ฆฌ๊ณ ์ด๊ฒ์ด AI์ ์ค๋ฆฌ์ ํฌ๋ช ์ฑ์ ์ด๋ป๊ฒ ๋ณํ์ํฌ ์ ์์๊น์? AI์ ์์ฌ๊ฒฐ์ ์ ๋ ์ ์ดํดํ๊ณ ๊ท์ ํ๋ ๋ฐ ๋์์ด ๋ ์ ์์๊น์?
AI๊ฐ ์ด ์ ๋๋ก ๋ฐ์ ํ๋ฉด ์ผ๋ถ ์ง์ ์ด ์ํฅ์ ๋ฐ์ ์๋ ์์ต๋๋ค. ์ฐ๋ฆฌ๋ ์ด๋ป๊ฒ ๋์ํด์ผ ํ ๊น์? ํธ๊ฒฌ์ ์ด๋ป๊ฒ ํ ๊น์? ์ด ๋ฐฉ๋ฒ์ด ํธ๊ฒฌ์ ์๋ณํ๊ณ ์์ ํ๋ ๋ฐ ๋์์ด ๋ ์ ์์๊น์?
๊ทธ๋ฆฌ๊ณ ์ค์ํ ์ง๋ฌธ์ด ํ๋ ๋ ์์ต๋๋ค. AI๊ฐ ์ฐ๋ฆฌ๋ณด๋ค ์ถ๋ก ์ ๋ ๋ฅ์ํด์ง๋ค๋ฉด, ์ฐ๋ฆฌ๊ฐ ํจ๊ป ์ผํ๋ ๋ฐฉ์์ด ์ด๋ป๊ฒ ๋ฐ๋๊น์?
๋ง์ง๋ง์ผ๋ก, ์ด ๋ฐฉ๋ฒ์ ์ธ๊ฐ์ด ์น์ธํ ํ๋ก์ธ์ค๋ฅผ ๋ฐ๋ฅด๊ธฐ ๋๋ฌธ์ AI์ ์ฐฝ์์ฑ์ด๋ ํ์ ์ ์ฌ๋ ฅ์ ์ ํํ ์ ์์๊น์?
๊ณ ๋ คํด์ผ ํ ์ฌํญ์ด ๋ง์ง๋ง ํ ๊ฐ์ง ํ์คํ ๊ฒ์ ์์ต๋๋ค: AI๋ ์ ์ ๋ ๋๋ํด์ง๊ณ ์์ผ๋ฉฐ, ์ด๋ฅผ ํ์ฉํ๋ ๊ฒ์ ์ฐ๋ฆฌ์ ๋ชซ์ด๋ผ๋ ๊ฒ์ ๋๋ค. #AI #OpenAI #ํ๋ก์ธ์ค๊ฐ๋ #MathWhizAI ๐ค๐ฉโ๐ซ๐ข
----
์ ์์ ์ผ๋ผ์ธ๋จผํธํ์ Jan Leike๊ฐ ์๋ ๊ฒ์ด ํฅ๋ฏธ๋กญ๊ตฐ์.
Jan Leike: https://twitter.com/janleike/status/1663977494058520576?fbclid=IwAR2uKE0yDJOc9V_2Ihh0AJY56TrFPs-BcAhG4kPBHmwR7fFaWrlmAtVq_nI
Leike๋ ์ด๊ฒ์ด ๊ธฐ๋ณธ์ ์ผ๋ก AI๊ฐ ์ฐ๋ฆฌ๊ฐ ์ํ๋ ๋ฐฉ์์ผ๋ก ์๋ํ๋๋ก ํ๋ '์ ๋ ฌ'์ด๋ผ๋ ์ธก๋ฉด์์ ํฐ ๋ด์ค๊ฐ ๋ ๊ฒ์ด๋ผ๊ณ ์๊ฐํ๋๊ตฐ์.
AI ์ ๊ณ์์๋ '์ ๋ ฌ ์ธ๊ธ'์ด๋ผ๊ณ ๋ถ๋ฆฌ๋ ๊ฒ, ์ฆ AI๋ฅผ ์ ๋ ฌํด์ ์๋ํ๊ฒ ๋ง๋ค๋ฉด AI์ ์ฑ๋ฅ์ด ๋จ์ด์ง ์ ์๋ค๋ ์ฐ๋ ค๊ฐ ์์์ต๋๋ค. ๊ทธ๋ฐ๋ฐ ์ด ์ฌ๋ก์ ๊ฒฝ์ฐ, AI๋ ์ค์ ๋ก ๋ฉด๋ฐํ ๊ฐ๋ ์ ๋ฐ์์ ๋ ๋ ์ ์ํํ๋ค๋ ๊ฒ์ด์ฃ .
Leike ์์ ๋ ์ด ์ฌ์ค์ ๋๋๋ ๋ด ๋๋ค. ๊ทธ๋ ๋ ํฐ AI ๋ชจ๋ธ์ด๋ ๋ค๋ฅธ ์์ญ์์๋ ๊ฐ์ ์ผ์ด ์ผ์ด๋ ์ง ํ์ ํ์ง ๋ชปํ์ง๋ง, ์ง๊ธ์ ํ์คํ ์ด ์์ด๋์ด์ ๋ํด ๋ ๊ฐ๋ฐฉ์ ์ด๊ตฐ์.
๊ทธ๋ ๋ํ ์ด๊ฒ์ด ์ ๋ ฌ ์ฐ๊ตฌ์ ํฐ ์๋ฏธ๋ฅผ ๊ฐ์ง ์ ์๋ค๊ณ ์๊ฐํฉ๋๋ค. AI์ ์๋ ๋ฐฉ์์ ๋ ์ ์ดํดํ ์ ์๋ค๋ฉด AI๊ฐ ๋ ์ ์๋ํ๋๋ก ํ ์ ์์ง ์์๊น์?
* PRM800K: A Process Supervision Dataset: https://github.com/openai/prm800k
----
* ํฌ์คํ ์์ฑ ๊ณผ์ : https://chat.openai.com/share/28921572-7b11-496c-a533-9898733d6577?fbclid=IwAR0qLLSN6H3aaiamCK-XFeZfCIa_wEXwBUBtlMjn9dN7h7F7UQH9f6l4qEQ
๊ณผ์ฐ ์ด๋ ๊ฒ ๊ธ์ฐ๊ธฐ๋ฅผ ํ๋ ๊ฒ์ ์ ๊ฐ ์ผ๋ค๊ณ ๋ณผ ์ ์๋ ๊ฒ์ธ์ง, ๊ณผ์ฐ ์ ๋ ์๋ฌธ์ ๋ด์ฉ์ ์ฝ์๋ค๊ณ ๋ณผ ์ ์๋ ๊ฒ์ธ์ง... ๊ธ์์.
๋ง์น ์๋ฌธ์ ์ฝ์ ๋ค๋ฅธ ์ฌ๋์ด ์ด ํฌ์คํ ์ ๋ณด๊ณ , ๊ทธ๊ฑธ ํตํด ๋๊ฐ ํ์ ํ๋ ๋๋์ด๋๊น์. ์ ๋ณด์ ํก์๋ผ๋๊ฒ ๋ ์ด๋ฐ ๋จ๊ณ๊ฐ ์๋ ๊ฒ์ด๊ธด ํ์ง๋ง... ๋ญ๊ฐ์ ๊ฐ๋ฆผ๊ธธ์ด๊ตฌ๋ ์ถ์ต๋๋ค. ์ฌ๊ธฐ์ ๋ ๋ค์ด์ ์ ๋๋ก ์ฝ๊ณ ์๊ฐํ ์ง ์๋๋ฉด ๊ฒ์ผ๋ฆ์ ํผ์ฐ๊ณ ๋ง์ง. ๊ทธ๋ฐ๋ฐ ์ ๋๋ก ์ฝ๊ณ ์๊ฐํ๋ค์ ์น๋๋ผ๋ ์๊ฐ์ด ์ง๋ ๋จ๋ ๊ฒ์ด ์ด๋ฐ ๊ฒ์ผ๋ฅธ ํฌ์คํ ๋๋น ๊ณผ์ฐ ๋ญ๊ฐ ๋ ์์ ๊ฒ์ธ๊ฐ... ์ด๊ฒ ๊ณ ๋ฏผ ์ง์ ์ ๋๋ค.
* ๋ญ๊ฐ ์ฐ์ฐํด์ ๊ฒฐ๊ตญ ๋ธ๋ก๊ทธ ์๋ฌธ์ ๋ค ์ฝ์์ต๋๋ค. ์ฝ๊ณ ๋์ ์ด ํฌ์คํ ์ ๋ณธ๋ฌธ์ ์ด๋ ๋คํ ์์ ์ ํ์ง ์๋ ๊ฒ๋ ์์ด๋ฌ๋ํฉ๋๋ค. ๋ ผ๋ฌธ๋ ๋๋ฌ๋ด์ผ์ฃ .
https://openai.com/research/improving-mathematical-reasoning-with-process-supervision?fbclid=IwAR3ICUhp_ALBMBUMOG_vqe1cqCMpLNTzm0yWkbIP2hbeeKX7OF-11_RCXQo
OpenAI๊ฐ ๋ ๋ค๋ฅธ ๋ํ๊ตฌ๋ฅผ ๋ง๋ จํ๊ตฐ์! ์ต์ข ์ ๋ต๋ฟ๋ง ์๋๋ผ ๊ฐ ์ ๋ต ๋จ๊ณ๋ง๋ค ๋ณด์์ ์ ๊ณตํ๋ ์ ๊ทผ์ผ๋ก ์ํ ๋ฌธ์ ํ์ด์์ ์๋ก์ด ๊ธฐ๋ก์ ์ธ์ด ๋ชจ๋ธ์ ํ๋ จ์์ผฐ ๋ค๊ณ ํฉ๋๋ค. ์ฐ๋ฆฌ๊ฐ ํ๊ต์์ ์ํ์ ๋ฐฐ์ฐ๋ ๋ฐฉ์๊ณผ ๋น์ทํ์ฃ ? ์ต์ข ์ ๋ต์ ๋งํ AI์ ๋ฑ์ ๋๋๋ฆฌ๋ ๋์ , ์ด์ ์ฌ๋ฐ๋ฅธ ๋จ๊ณ๋ฅผ ๋ฐ์ ๋๋ง๋ค ๋ณด์์ ์ ๊ณตํฉ๋๋ค. ๋ง์น ์ํ ์ ์๋์ด ์ต์ข ์ ๋ต๋ฟ๋ง ์๋๋ผ ํ์ต ๊ณผ์ ๊น์ง ์ ๊ฒฝ ์จ์ฃผ๋ ๊ฒ์ฒ๋ผ์. ์ด๋ฅผ '๊ณผ์ ๊ฐ๋ (Process Supervision)'์ด๋ผ๊ณ ๋ถ๋ฅด๋๋ฐ, ์ด๋ ํ๊ธฐ์ ์ธ ๋ณํ์ ๋๋ค. ๐งฎ๐ก
* ๋ ผ๋ฌธ์ ์ ๋ชฉ์ "Letโs Verify Step by Step"์ ๋๋ค. (์๋ ์ CoT - '์๊ฐ์ ์ฌ์ฌ' ๋ถ์ ์ผ์ผ์ผฐ๋ ๋ง๋ฒ์ ํํ Step by Step ์ ๋๋ค)
https://cdn.openai.com/improving-mathematical-reasoning-with-process-supervision/Lets_Verify_Step_by_Step.pdf?fbclid=IwAR1LojRtgWmMYERWatOlHsD_9AAxalBPDd2i9WiDfRwV_F_Scm3eT2IUB1k
์ด๊ฒ์ ๋จ์ํ ์ํ ๋ฌธ์ ๋ฅผ ๋ ์ ํธ๋ ๊ฒ๋ง์ด ์๋๋๋ค. ํ๋ก์ธ์ค ๊ฐ๋ ์ ๋ชจ๋ธ์ด ์ฌ๋์ด ์น์ธํ ํ๋ก์ธ์ค๋ฅผ ๋ฐ๋ผ ์ฐ๋ฆฌ์ ๋ ๋น์ทํ๊ฒ ์๊ฐํ๋๋ก ๋์์ค๋๋ค. ๋ง์น ์ธ๊ณต์ง๋ฅ์ ๋ง์์ ๋ค์ฌ๋ค๋ณด๊ณ ์๋ํ๋ ๊ณผ์ ์ ์ง์ผ๋ณด๋ ๊ฒ๊ณผ ๋น์ทํฉ๋๋ค. ์ด ๋ฐฉ๋ฒ์ ๋งค์ฐ ์ ๋ขฐํ ์ ์๋ ๊ฒ์ผ๋ก ์ ์ฆ๋์์ผ๋ฉฐ, AI๋ฅผ ๋ ์ ์ดํดํ๊ณ ์ฐ๋ฆฌ์ ๋ชฉํ์ ๋ถํฉํ๋๋ก ๋ง๋ค ์๋ ์์ต๋๋ค.
์ด๊ฒ์ AI์ ์๋๋ฅผ ๋ฆ์ถ๋ ๊ฒ์ด ์๋๋ผ ์คํ๋ ค ๋ ๋์ ์ฑ๋ฅ์ ์ ๊ณตํฉ๋๋ค! ์์ ๊ณผ ์ฑ๋ฅ ๋ชจ๋์์ ์์์ธ ์ ์ด์ฃ .
๋งค์ฐ ๊น๋ค๋ก์ด ์ํ ๋ฌธ์ ๋ฅผ ํ ์คํธํ ๊ฒฐ๊ณผ, '๊ณผ์ ๊ฐ๋ ํ' AI๊ฐ '๊ฒฐ๊ณผ ๊ฐ๋ ํ' AI๋ณด๋ค ์ ๋ฐ์ ์ผ๋ก ๋ ๋ฐ์ด๋ ์ฑ๋ฅ์ ๋ณด์์ต๋๋ค. ์ฌ์ง์ด ๋๋ถ๋ถ์ ์๋๊ฐ ์คํจ๋ก ๋์๊ฐ ์ด๋ ค์ด ์ผ๊ฐํ ๋ฌธ์ ์ ๋ํ ์ ํจํ ํด๊ฒฐ์ฑ ์ ์ฐพ์๋ด๊ธฐ๋ ํ์ต๋๋ค! ๐ฎ
ํ์ง๋ง ์ด ์ฑ๊ณผ๋ฅผ ์นญ์ฐฌํ๋๋ฐ๋ง ๊ทธ์น๋ ๊ฒ์ด ์๋๋ผ ์ง์งํ๊ฒ ๊ณ ๋ฏผํด์ผ ํ ๋ถ๋ถ์ด ์์ต๋๋ค. ๐ค
์ด ๋ฐฉ๋ฒ์ ๋ค๋ฅธ ๋ฌธ์ ํด๊ฒฐ ์์ ์์ AI๋ฅผ ํ๋ จํ๋ ๋ฐ ์ฌ์ฉํ ์ ์์๊น์? ๊ทธ๋ฆฌ๊ณ ์ด๊ฒ์ด AI์ ์ค๋ฆฌ์ ํฌ๋ช ์ฑ์ ์ด๋ป๊ฒ ๋ณํ์ํฌ ์ ์์๊น์? AI์ ์์ฌ๊ฒฐ์ ์ ๋ ์ ์ดํดํ๊ณ ๊ท์ ํ๋ ๋ฐ ๋์์ด ๋ ์ ์์๊น์?
AI๊ฐ ์ด ์ ๋๋ก ๋ฐ์ ํ๋ฉด ์ผ๋ถ ์ง์ ์ด ์ํฅ์ ๋ฐ์ ์๋ ์์ต๋๋ค. ์ฐ๋ฆฌ๋ ์ด๋ป๊ฒ ๋์ํด์ผ ํ ๊น์? ํธ๊ฒฌ์ ์ด๋ป๊ฒ ํ ๊น์? ์ด ๋ฐฉ๋ฒ์ด ํธ๊ฒฌ์ ์๋ณํ๊ณ ์์ ํ๋ ๋ฐ ๋์์ด ๋ ์ ์์๊น์?
๊ทธ๋ฆฌ๊ณ ์ค์ํ ์ง๋ฌธ์ด ํ๋ ๋ ์์ต๋๋ค. AI๊ฐ ์ฐ๋ฆฌ๋ณด๋ค ์ถ๋ก ์ ๋ ๋ฅ์ํด์ง๋ค๋ฉด, ์ฐ๋ฆฌ๊ฐ ํจ๊ป ์ผํ๋ ๋ฐฉ์์ด ์ด๋ป๊ฒ ๋ฐ๋๊น์?
๋ง์ง๋ง์ผ๋ก, ์ด ๋ฐฉ๋ฒ์ ์ธ๊ฐ์ด ์น์ธํ ํ๋ก์ธ์ค๋ฅผ ๋ฐ๋ฅด๊ธฐ ๋๋ฌธ์ AI์ ์ฐฝ์์ฑ์ด๋ ํ์ ์ ์ฌ๋ ฅ์ ์ ํํ ์ ์์๊น์?
๊ณ ๋ คํด์ผ ํ ์ฌํญ์ด ๋ง์ง๋ง ํ ๊ฐ์ง ํ์คํ ๊ฒ์ ์์ต๋๋ค: AI๋ ์ ์ ๋ ๋๋ํด์ง๊ณ ์์ผ๋ฉฐ, ์ด๋ฅผ ํ์ฉํ๋ ๊ฒ์ ์ฐ๋ฆฌ์ ๋ชซ์ด๋ผ๋ ๊ฒ์ ๋๋ค. #AI #OpenAI #ํ๋ก์ธ์ค๊ฐ๋ #MathWhizAI ๐ค๐ฉโ๐ซ๐ข
----
์ ์์ ์ผ๋ผ์ธ๋จผํธํ์ Jan Leike๊ฐ ์๋ ๊ฒ์ด ํฅ๋ฏธ๋กญ๊ตฐ์.
Jan Leike: https://twitter.com/janleike/status/1663977494058520576?fbclid=IwAR2uKE0yDJOc9V_2Ihh0AJY56TrFPs-BcAhG4kPBHmwR7fFaWrlmAtVq_nI
Leike๋ ์ด๊ฒ์ด ๊ธฐ๋ณธ์ ์ผ๋ก AI๊ฐ ์ฐ๋ฆฌ๊ฐ ์ํ๋ ๋ฐฉ์์ผ๋ก ์๋ํ๋๋ก ํ๋ '์ ๋ ฌ'์ด๋ผ๋ ์ธก๋ฉด์์ ํฐ ๋ด์ค๊ฐ ๋ ๊ฒ์ด๋ผ๊ณ ์๊ฐํ๋๊ตฐ์.
AI ์ ๊ณ์์๋ '์ ๋ ฌ ์ธ๊ธ'์ด๋ผ๊ณ ๋ถ๋ฆฌ๋ ๊ฒ, ์ฆ AI๋ฅผ ์ ๋ ฌํด์ ์๋ํ๊ฒ ๋ง๋ค๋ฉด AI์ ์ฑ๋ฅ์ด ๋จ์ด์ง ์ ์๋ค๋ ์ฐ๋ ค๊ฐ ์์์ต๋๋ค. ๊ทธ๋ฐ๋ฐ ์ด ์ฌ๋ก์ ๊ฒฝ์ฐ, AI๋ ์ค์ ๋ก ๋ฉด๋ฐํ ๊ฐ๋ ์ ๋ฐ์์ ๋ ๋ ์ ์ํํ๋ค๋ ๊ฒ์ด์ฃ .
Leike ์์ ๋ ์ด ์ฌ์ค์ ๋๋๋ ๋ด ๋๋ค. ๊ทธ๋ ๋ ํฐ AI ๋ชจ๋ธ์ด๋ ๋ค๋ฅธ ์์ญ์์๋ ๊ฐ์ ์ผ์ด ์ผ์ด๋ ์ง ํ์ ํ์ง ๋ชปํ์ง๋ง, ์ง๊ธ์ ํ์คํ ์ด ์์ด๋์ด์ ๋ํด ๋ ๊ฐ๋ฐฉ์ ์ด๊ตฐ์.
๊ทธ๋ ๋ํ ์ด๊ฒ์ด ์ ๋ ฌ ์ฐ๊ตฌ์ ํฐ ์๋ฏธ๋ฅผ ๊ฐ์ง ์ ์๋ค๊ณ ์๊ฐํฉ๋๋ค. AI์ ์๋ ๋ฐฉ์์ ๋ ์ ์ดํดํ ์ ์๋ค๋ฉด AI๊ฐ ๋ ์ ์๋ํ๋๋ก ํ ์ ์์ง ์์๊น์?
* PRM800K: A Process Supervision Dataset: https://github.com/openai/prm800k
----
* ํฌ์คํ ์์ฑ ๊ณผ์ : https://chat.openai.com/share/28921572-7b11-496c-a533-9898733d6577?fbclid=IwAR0qLLSN6H3aaiamCK-XFeZfCIa_wEXwBUBtlMjn9dN7h7F7UQH9f6l4qEQ
๊ณผ์ฐ ์ด๋ ๊ฒ ๊ธ์ฐ๊ธฐ๋ฅผ ํ๋ ๊ฒ์ ์ ๊ฐ ์ผ๋ค๊ณ ๋ณผ ์ ์๋ ๊ฒ์ธ์ง, ๊ณผ์ฐ ์ ๋ ์๋ฌธ์ ๋ด์ฉ์ ์ฝ์๋ค๊ณ ๋ณผ ์ ์๋ ๊ฒ์ธ์ง... ๊ธ์์.
๋ง์น ์๋ฌธ์ ์ฝ์ ๋ค๋ฅธ ์ฌ๋์ด ์ด ํฌ์คํ ์ ๋ณด๊ณ , ๊ทธ๊ฑธ ํตํด ๋๊ฐ ํ์ ํ๋ ๋๋์ด๋๊น์. ์ ๋ณด์ ํก์๋ผ๋๊ฒ ๋ ์ด๋ฐ ๋จ๊ณ๊ฐ ์๋ ๊ฒ์ด๊ธด ํ์ง๋ง... ๋ญ๊ฐ์ ๊ฐ๋ฆผ๊ธธ์ด๊ตฌ๋ ์ถ์ต๋๋ค. ์ฌ๊ธฐ์ ๋ ๋ค์ด์ ์ ๋๋ก ์ฝ๊ณ ์๊ฐํ ์ง ์๋๋ฉด ๊ฒ์ผ๋ฆ์ ํผ์ฐ๊ณ ๋ง์ง. ๊ทธ๋ฐ๋ฐ ์ ๋๋ก ์ฝ๊ณ ์๊ฐํ๋ค์ ์น๋๋ผ๋ ์๊ฐ์ด ์ง๋ ๋จ๋ ๊ฒ์ด ์ด๋ฐ ๊ฒ์ผ๋ฅธ ํฌ์คํ ๋๋น ๊ณผ์ฐ ๋ญ๊ฐ ๋ ์์ ๊ฒ์ธ๊ฐ... ์ด๊ฒ ๊ณ ๋ฏผ ์ง์ ์ ๋๋ค.
* ๋ญ๊ฐ ์ฐ์ฐํด์ ๊ฒฐ๊ตญ ๋ธ๋ก๊ทธ ์๋ฌธ์ ๋ค ์ฝ์์ต๋๋ค. ์ฝ๊ณ ๋์ ์ด ํฌ์คํ ์ ๋ณธ๋ฌธ์ ์ด๋ ๋คํ ์์ ์ ํ์ง ์๋ ๊ฒ๋ ์์ด๋ฌ๋ํฉ๋๋ค. ๋ ผ๋ฌธ๋ ๋๋ฌ๋ด์ผ์ฃ .
Openai
Improving mathematical reasoning with process supervision
We've trained a model to achieve a new state-of-the-art in mathematical problem solving by rewarding each correct step of reasoning (โprocess supervisionโ) instead of simply rewarding the correct final answer (โoutcome supervisionโ). In addition to boostingโฆ
What if we apply this methodology into other areas? Maybe at some point, human need to learn from AI about how to learn...
Japan Goes All In: Copyright Doesnโt Apply To AI Training
์ผ๋ณธ ์ ๋ถ๋ ์ธ๊ณต์ง๋ฅ ํ์ต์ ์ฌ์ฉ๋๋ ๋ฐ์ดํฐ์ ๋ํด์ ์ ์๊ถ์ ๊ฐ์ ํ์ง ์๊ฒ ๋ต๋๋ค.
"์ผ๋ณธ ์ ๋ถ๋ ์ต๊ทผ ๋๋๊ฒ๋ AI ํ์ต์ ์ฌ์ฉ๋๋ ๋ฐ์ดํฐ์ ๋ํด ์ ์๊ถ์ ๊ฐ์ ํ์ง ์๊ฒ ๋ค๋ ์ ์ฅ์ ์ฌํ์ธํ์ต๋๋ค. ์ด ์ ์ฑ ์ "๋น์๋ฆฌ ๋ชฉ์ ์ด๋ ์์ ์ ๋ชฉ์ ์ด๋ , ๋ณต์ ๊ฐ ์๋ ๋ค๋ฅธ ํ์์ด๋ , ๋ถ๋ฒ ์ฌ์ดํธ ๋ฑ์์ ์ป์ ์ฝํ ์ธ ์ด๋ ์๊ด์์ด" AI๊ฐ ๋ชจ๋ ๋ฐ์ดํฐ๋ฅผ ์ฌ์ฉํ ์ ์๋๋ก ํ์ฉํฉ๋๋ค. ์ผ๋ณธ ๋ฌธ๋ถ๊ณผํ์ฑ ์ฅ๊ด์ธ ๋๊ฐ์ค์นด ์ผ์ด์ฝ๋ ํ์ง ํ์์์ ์ผ๋ณธ ๋ฒ๋ฅ ์ด AI ๋ฐ์ดํฐ ์ธํธ์ ์ฌ์ฉ๋ ์ ์๊ถ์ด ์๋ ์๋ฃ๋ฅผ ๋ณดํธํ์ง ์์ ๊ฒ์ด๋ผ๊ณ ๋งํ๋ฉฐ ๋๋ดํ ์ ์ฅ์ ํ์ธํ์ต๋๋ค."
๊ทธ๋ฐ๋ฐ ์ด์ ๋์ํ๋ ๊ตญ๋ด ๊ธฐ์ฌ๋ ๋ชป์ฐพ๊ฒ ๋ค์. (์๋๋ฐ ์ ๊ฐ ์ฐพ์ง ๋ชปํ๋ ๊ฑธ ์๋ ์์ต๋๋ค.)
https://technomancers.ai/japan-goes-all-in-copyright-doesnt-apply-to-ai-training/?fbclid=IwAR0jWJD6EvO9eE6hTdjqUBOuGkTXVYIVZ1iuP_cLgqj7_Sz2xs8K4tKgyTM
Huge decision for Japan. Smart goverment.
์ผ๋ณธ ์ ๋ถ๋ ์ธ๊ณต์ง๋ฅ ํ์ต์ ์ฌ์ฉ๋๋ ๋ฐ์ดํฐ์ ๋ํด์ ์ ์๊ถ์ ๊ฐ์ ํ์ง ์๊ฒ ๋ต๋๋ค.
"์ผ๋ณธ ์ ๋ถ๋ ์ต๊ทผ ๋๋๊ฒ๋ AI ํ์ต์ ์ฌ์ฉ๋๋ ๋ฐ์ดํฐ์ ๋ํด ์ ์๊ถ์ ๊ฐ์ ํ์ง ์๊ฒ ๋ค๋ ์ ์ฅ์ ์ฌํ์ธํ์ต๋๋ค. ์ด ์ ์ฑ ์ "๋น์๋ฆฌ ๋ชฉ์ ์ด๋ ์์ ์ ๋ชฉ์ ์ด๋ , ๋ณต์ ๊ฐ ์๋ ๋ค๋ฅธ ํ์์ด๋ , ๋ถ๋ฒ ์ฌ์ดํธ ๋ฑ์์ ์ป์ ์ฝํ ์ธ ์ด๋ ์๊ด์์ด" AI๊ฐ ๋ชจ๋ ๋ฐ์ดํฐ๋ฅผ ์ฌ์ฉํ ์ ์๋๋ก ํ์ฉํฉ๋๋ค. ์ผ๋ณธ ๋ฌธ๋ถ๊ณผํ์ฑ ์ฅ๊ด์ธ ๋๊ฐ์ค์นด ์ผ์ด์ฝ๋ ํ์ง ํ์์์ ์ผ๋ณธ ๋ฒ๋ฅ ์ด AI ๋ฐ์ดํฐ ์ธํธ์ ์ฌ์ฉ๋ ์ ์๊ถ์ด ์๋ ์๋ฃ๋ฅผ ๋ณดํธํ์ง ์์ ๊ฒ์ด๋ผ๊ณ ๋งํ๋ฉฐ ๋๋ดํ ์ ์ฅ์ ํ์ธํ์ต๋๋ค."
๊ทธ๋ฐ๋ฐ ์ด์ ๋์ํ๋ ๊ตญ๋ด ๊ธฐ์ฌ๋ ๋ชป์ฐพ๊ฒ ๋ค์. (์๋๋ฐ ์ ๊ฐ ์ฐพ์ง ๋ชปํ๋ ๊ฑธ ์๋ ์์ต๋๋ค.)
https://technomancers.ai/japan-goes-all-in-copyright-doesnt-apply-to-ai-training/?fbclid=IwAR0jWJD6EvO9eE6hTdjqUBOuGkTXVYIVZ1iuP_cLgqj7_Sz2xs8K4tKgyTM
Huge decision for Japan. Smart goverment.
Technomancers.ai
Japan Goes All In: Copyright Doesn't Apply To AI Training
Training data can be used "regardless of whether it is for non-profit or commercial purposes, whether it is an act other than reproduction, or whether it is content obtained from illegal sites or otherwise."
Continuous Learning_Startup & Investment
honeycomb.io/blog/hard-stuff-nobody-talks-about-llm?fbclid=IwAR1pDrPdN7Vu9PSEwK6aD_lWCZMoZpbf5cQjI0nZ1MDWQfJUNO8ZUM27G74_aem_th_AUxsqnFXcjp2E4Hl4UUDPEp6EK_W_tS9Vkgp9aXhW0Kq5cKDefW90jh2ARchnFjpies&mibextid=Zxz2cZ
Builderโs point of view about LLM.
The article discusses the challenges of building products with large language models (LLMs). These models are trained on massive datasets of text and code, and can be used for a variety of tasks, such as generating text, translating languages, and writing different kinds of creative content. However, building products with LLMs is not without its challenges. One challenge is that LLMs are often very large and complex, which can make them difficult to train and deploy. Another challenge is that LLMs can be difficult to control, and can sometimes generate output that is offensive or harmful. Finally, LLMs can be expensive to use, and there is a risk that they could be used for malicious purposes.
Despite these challenges, the author believes that LLMs have the potential to revolutionize the way we build products. He argues that LLMs can be used to create more personalized and engaging experiences for users, and that they can be used to solve complex problems that were previously intractable. He concludes by calling for more research and development into the use of LLMs in products.
Here are some of the specific challenges mentioned in the article:
Training and deploying LLMs is expensive and time-consuming. LLMs require massive datasets of text and code to train, and can take weeks or even months to train. Once they are trained, they can be difficult to deploy, as they require specialized hardware and software.
LLMs can be difficult to control. LLMs are trained on massive datasets of text, and can sometimes generate output that is offensive or harmful. This can be a challenge for developers, who need to find ways to control the output of LLMs.
LLMs can be expensive to use. LLMs require specialized hardware and software, which can be expensive to purchase and maintain. Additionally, LLMs can be expensive to train, as they require massive datasets of text and code.
There is a risk that LLMs could be used for malicious purposes. LLMs are powerful tools, and could be used for malicious purposes, such as generating fake news or creating deepfakes. It is important to be aware of these risks when developing products with LLMs.
The article discusses the challenges of building products with large language models (LLMs). These models are trained on massive datasets of text and code, and can be used for a variety of tasks, such as generating text, translating languages, and writing different kinds of creative content. However, building products with LLMs is not without its challenges. One challenge is that LLMs are often very large and complex, which can make them difficult to train and deploy. Another challenge is that LLMs can be difficult to control, and can sometimes generate output that is offensive or harmful. Finally, LLMs can be expensive to use, and there is a risk that they could be used for malicious purposes.
Despite these challenges, the author believes that LLMs have the potential to revolutionize the way we build products. He argues that LLMs can be used to create more personalized and engaging experiences for users, and that they can be used to solve complex problems that were previously intractable. He concludes by calling for more research and development into the use of LLMs in products.
Here are some of the specific challenges mentioned in the article:
Training and deploying LLMs is expensive and time-consuming. LLMs require massive datasets of text and code to train, and can take weeks or even months to train. Once they are trained, they can be difficult to deploy, as they require specialized hardware and software.
LLMs can be difficult to control. LLMs are trained on massive datasets of text, and can sometimes generate output that is offensive or harmful. This can be a challenge for developers, who need to find ways to control the output of LLMs.
LLMs can be expensive to use. LLMs require specialized hardware and software, which can be expensive to purchase and maintain. Additionally, LLMs can be expensive to train, as they require massive datasets of text and code.
There is a risk that LLMs could be used for malicious purposes. LLMs are powerful tools, and could be used for malicious purposes, such as generating fake news or creating deepfakes. It is important to be aware of these risks when developing products with LLMs.
10 common mistakes startups make that result in slower growth:
1. Focusing on middle of the funnel when the bottleneck is top of the funnel. Look at reps calendars. Are they busy with multiple customer calls every day? If not, you have a demand (and/or p/m fit) bottleneck.
2. Mass email sequencing people who have never heard of you or your company. You need to be more creative with outbound.
3. Incentivizing the wrong behavior. Most commonly, top of funnel metrics like MQLโs and SQLโs that produce low conversion rates and low ACVโs. โShow me the incentive and Iโll show you the outcome.โ - Charlie Munger
4. Quarterly quotas. Almost every startup should be on monthly quotas. The exception being high 6 figure ACVs with >90 day sales cycles. The consequence of quarterly quotas is lost urgency 2 out of 3 months and less revenue.
5. Underinvesting or investing too late in revenue operations.
6. Underinvesting or investing too late in implementation. The first 30-60 days are the most critical in determining the long term success of a customer. Do whatever it takes to deliver a delightful first impression.
7. Underinvesting or investing too late in revenue expansion in the customer base.
8. Never meeting customers in person. Founders and revenue leaders should be doing this regularly (yes, even if your customers are small).
9. Over-indexing on domain experience when hiring a VP of Sales. Or, less common, but an even worse mistake, hiring from a big company.
10. Allowing poor performers, especially those in leadership positions, to remain at the business for too long.
https://www.linkedin.com/posts/sam-blond-791026b_10-common-mistakes-startups-make-that-result-activity-7070149376043573250-QLTB?utm_source=share&utm_medium=member_desktop
1. Focusing on middle of the funnel when the bottleneck is top of the funnel. Look at reps calendars. Are they busy with multiple customer calls every day? If not, you have a demand (and/or p/m fit) bottleneck.
2. Mass email sequencing people who have never heard of you or your company. You need to be more creative with outbound.
3. Incentivizing the wrong behavior. Most commonly, top of funnel metrics like MQLโs and SQLโs that produce low conversion rates and low ACVโs. โShow me the incentive and Iโll show you the outcome.โ - Charlie Munger
4. Quarterly quotas. Almost every startup should be on monthly quotas. The exception being high 6 figure ACVs with >90 day sales cycles. The consequence of quarterly quotas is lost urgency 2 out of 3 months and less revenue.
5. Underinvesting or investing too late in revenue operations.
6. Underinvesting or investing too late in implementation. The first 30-60 days are the most critical in determining the long term success of a customer. Do whatever it takes to deliver a delightful first impression.
7. Underinvesting or investing too late in revenue expansion in the customer base.
8. Never meeting customers in person. Founders and revenue leaders should be doing this regularly (yes, even if your customers are small).
9. Over-indexing on domain experience when hiring a VP of Sales. Or, less common, but an even worse mistake, hiring from a big company.
10. Allowing poor performers, especially those in leadership positions, to remain at the business for too long.
https://www.linkedin.com/posts/sam-blond-791026b_10-common-mistakes-startups-make-that-result-activity-7070149376043573250-QLTB?utm_source=share&utm_medium=member_desktop
Linkedin
Sam Blond on LinkedIn: 10 common mistakes startups make that result in slower growth:
1โฆ | 14 comments
1โฆ | 14 comments
10 common mistakes startups make that result in slower growth:
1. Focusing on middle of the funnel when the bottleneck is top of the funnel. Look at repsโฆ | 14 comments on LinkedIn
1. Focusing on middle of the funnel when the bottleneck is top of the funnel. Look at repsโฆ | 14 comments on LinkedIn