#claude2 — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #claude2, aggregated by home.social.
-
How can #largeLanguageModels scale up development of and insight from #ethics dilemmas?
Researchers had #LLMs make "50 scenarios and 400 unique test items" and collected responses from humans, #GPT4, and #Claude2.
- harm that was necessary to reduce overall harm deemed worse (and more intentional) than harm that was merely a side effect
- same for evitable harm (vs. inevitable) harm
- no effects of harm that required action (vs. omission) -
Hey everyone, I came across something interesting on Poe! They have a feature called Geminis_prompt which is an AI prompt generator called Gemini. It allows you to ask questions, get instant answers, and even have conversations with various AI models like ChatGPT, GPT-4, Claude 2, and DALLE 3. If you're curious and want to explore more, check out Poe's website. It's definitely worth a look! #Poe #AI #Gemini #ChatGPT #GPT-4 #Claude2 #DALLE3
Check this out on Poe: https://poe.com/Geminis_prompt -
Claude 3 seems to detect when it is being tested, sparking AI buzz online - Enlarge (credit: Getty Images)
On Monday, Anthropic prompt eng... - https://arstechnica.com/?p=2007736 #largelanguagemodels #margaretmitchell #machinelearning #textsynthesis #aiemotions #aifeelings #anthropic #aiethics #chatgpt #chatgtp #claude2 #claude3 #biz #claude #openai #sonnet #haiku #opus #ai
-
The AI wars heat up with Claude 3, claimed to have “near-human” abilities - Enlarge / The Anthropic Claude 3 logo. (credit: Anthropic)
On ... - https://arstechnica.com/?p=2007511 #machinelearning #anthropic #chatgpt #claude2 #claude3 #biz #claude #openai #sonnet #haiku #opus #ai
-
Politische Chats mit Mensch oder Maschine?
Schon jetzt können Menschen in mehr als der Hälfte aller Fälle nicht mehr unterscheiden, mit wem sie chatten https://www.telepolis.de/features/KI-Chatbots-in-der-politischen-Kommunikation-taeuschend-echt-9644300.html
#ai #ki #llm #chatbot #fakenews #GPT4 #Llama2 #Claude2 -
🧠 Un confronto tra output di diversi #LLM: #GPT4, GPT-3.5, #Gemini Plus, #Claude2, #Llama2 70b, #Mixtral 8x7b (input identico).
⚙️ Il task è molto semplice: l'analisi di una recensione. Le risposte sono molto simili: le sfumature nell'estrazione dei topic derivano da diverse letture del contesto, ma hanno tutti senso.
🦾 Per operazioni semplici e ricorrenti, i modelli open source eseguiti localmente o su istanze private possono essere una risposta di valore.
-
AI poisoning could turn open models into destructive “sleeper agents,” says Anthropic - Enlarge (credit: Benj Edwards | Getty Images)
Imagine download... - https://arstechnica.com/?p=1995975 #largelanguagemodels #promptinjections #sleeperagents #llmsecurity #aisecurity #anthropic #chatgpt #chatgtp #claude2 #biz #claude #llm #ai
-
Tales from the jar side: Claude AI in the Spring Framework, a viral video on YouTube plagiarism, some upcoming milestones, and the usual assortment of silly tweets, toots, and skeets
#java #springboot #plagiarism #claude2 #claudeai
https://open.substack.com/pub/kenkousen/p/tales-from-the-jar-side-claude-in?r=2dwq5&utm_campaign=post&utm_medium=web -
This week's Tales from the jar side, about how Google's Bard is stupid, Anthropic's Claude is better but can't do math, this week in Elon, and the usual silly tweets, toots, and skeets
https://open.substack.com/pub/kenkousen/p/tales-from-the-jar-side-google-bard?r=2dwq5&utm_campaign=post&utm_medium=web
#java #claude2 #openai #bard #elon -
-
Me, whenever someone puts some #StochasticParrot-related or -generated report or proposal on my desk.
-
Elon Musk’s new AI model doesn’t shy from questions about cocaine and orgies - Enlarge (credit: Getty Images | Benj Edwards)
On Saturday, Elo... - https://arstechnica.com/?p=1981276 #largelanguagemodels #largelanguagemodel #machinelearning #culturewars #anthropic #elonmusk #chatgpt #chatgtp #claude2 #twitter #biz #llama2 #grok #meta #woke #x.ai #ai
-
Stanford researchers challenge OpenAI, others on AI transparency in new report - Enlarge (credit: Getty Images / Benj Edwards)
On Wednesday, St... - https://arstechnica.com/?p=1977869 #foundationmodeltransparencyindex #largelanguagemodels #machinelearning #aitransparency #rishibommasani #amazontitan #percyliang #anthropic #aiethics #stanford #chatgpt #chatgtp #claude2 #biz #amazon #google #openai #gpt-3 #gpt-4 #palm2 #tedai #meta #ai
-
#AI #GenerativeAI #Lyrics #Anthropic #Claude2 #UniversalMusic #Copyright #IP: "Universal Music has sued artificial intelligence startup Anthropic over “systematic and widespread infringement of their copyrighted song lyrics,” per a filing Wednesday in a Tennessee federal court.
One example from the lawsuit: When a user asks Anthropic’s AI chatbot Claude about the lyrics to the song “Roar” by Katy Perry, it generates an “almost identical copy of those lyrics,” violating the rights of Concord, the copyright owner, per the filing. The lawsuit also named Gloria Gaynor’s “I Will Survive” as an example of Anthropic’s alleged copyright infringement, as Universal owns the rights to its lyrics.
“In the process of building and operating AI models, Anthropic unlawfully copies and disseminates vast amounts of copyrighted works,” the lawsuit stated, later going on to add, “Just like the developers of other technologies that have come before, from the printing press to the copy machine to the web-crawler, AI companies must follow the law.”"
-
Amazon інвестує $4 млрд в Anthropic — розробника конкурентного чатбота для ChatGPT https://itc.ua/ua/novini/amazon-investuye-4-mlrd-v-anthropic-rozrobnyka-konkurentnogo-chatbota-dlya-chatgpt/ #Великімовнімоделі #Штучнийінтелект #ГенеративнийШІ #Нейромережі #Технології #ChatGPT #Claude2 #Новини #Amazon
-
Claude 2 vs ChatGPT: i vantaggi che devi conoscere.
Claude 2 è il nuovo contendente nella scena delle IA generative, pronto a sfidare ChatGPT.👉 https://gomoot.com/claude-2-vs-chatgpt-i-vantaggi-che-devi-conoscere/
-
The New York Times prohibits AI vendors from devouring its content - Enlarge (credit: Benj Edwards / Getty Images)
In early August,... - https://arstechnica.com/?p=1960621 #largelanguagemodels #machinelearning #thenewyorktimes #googlebard #journalism #anthropic #aiethics #chatgtp #claude2 #biz #llama2 #openai #palm2 #tech #meta #ai
-
🧠 Interessante studio condotto da Stanford, Berkeley e Samaya #AI: su finestre di contesto grandi le performance dei #LLM su diversi task sono migliori nella parte iniziale e finale, rispetto a quella centrale.
🧐 E più il contesto aumenta, più le performance diminuiscono, con un elevato degrado iniziale che poi tende a stabilizzarsi.✨ Non c'è dubbio sul fatto che la vettorializzazione del contesto darà sempre prestazioni migliori.
🔗 Lo studio: https://arxiv.org/pdf/2307.03172.pdf
-
I tested #claude2 and it failed the Balrog Test (miserably).
The Balrog Test goes like this: ask the AI if balrogs have wings. The answer is complicated.
They don't have wings made of flesh and feathers (although they could if they wished because they're Maiar, capable of fashioning any kind of body they want).
The Balrog of Moria (which attacked the Fellowship of the Ring and killed Gandalf) surrounded itself with a darkness that Tolkien described as "a great shadow" and it could cause that darkness/shadow to take the shape of "vast wings". It was these wings (made of the darkness, not flesh and blood) that extended from wall to wall.
So, no physical wings, but not metaphorical, either.
If an AI can explain all that (and none that I have tested can), then it passes the Balrog Test.
Claude2 made up a completely bogus letter by J.R.R. Tolkien, falsely claiming that HE said balrogs have no wings. It assigned the no. 144 to the bogus letter. When I say "made up", I mean it quoted extensively from this non-existent letter.
Even #chatgpt did better than that.
-
I guarantee you we'll see a bajillion blog posts telling us how great and reliable Claude-2 is with no mention whatsoever of these disclaimers (or they'll downplay the importance of the disclaimers).
-
@itsthejoker it's really tough tbh. I found #chatgpt immensely helpful though. Worth testing out #claude2 as well as it has huge token context which allows it to ingest your whole project so it can provide more accurate suggestions.
The time is now though! It has never been this easy to onboard on a new programming language :)
-
I've been playing with Anthropic's Claude2 chatbot. It can code in R, including the tidyverse, data.table, and base R. I asked which it prefers 😀
You can access Claude here https://claude.ai/
#rstats #GenerativeAI, #LLM #LLMs #Anthropic #Claude #Claude2
-
Nice overview video of Claude 2.
#AI #Claude2 #LLM #GenerativeAI
https://youtu.be/o1SR_ecQ88A -
#Claude2 so far is better than Bard and on par with ChatGPT.
-
New ChatGPT rival, Claude 2, launches for open beta testing - Enlarge (credit: Anthropic)
On Tuesday, Anthropic introduced C... - https://arstechnica.com/?p=1952822 #largelanguagemodels #machinelearning #anthropic #chatgpt #claude2 #biz #claude #openai #ai
