fleszarjacek
A
Today, neither ChatGPT nor Perplexity or Bard can really replace Google Search. However, it's not hard to imagine an application that functions as a chatbot and a web navigator👈
07:51 AM - Feb 15, 2024
0
0
0
fleszarjacek
A
Al have ability to reason, and generate functioning code in languages like Python, SQL, and R, they can provide impressive value with Data analysis. But can they replace data analysts? #artificialwisdom
12:18 PM - Jan 03, 2024
0
0
0
fleszarjacek
A
It's never been more important to come together and collaborate on open-source LLMs and AI,

Come network with other developers and researchers and party in SF on Dec 7th🙋
02:40 PM - Nov 28, 2023
0
0
0
fleszarjacek
A
A $10M prize was just announced for anyone who releases a public model that does well on math olympiad questions - so thre is a lot of impetus to get better.
#artificialintelligencentelligence
03:36 AM - Nov 28, 2023
0
0
0
fleszarjacek
A
Agency and consciousness evolved in carbon life forms over millions of years because of environmental pressures. LLMs face no such fitness functions or evolutionary pressures.
#artificialintelligencentelligence
04:38 AM - Nov 26, 2023
Avatar
0
0
1
fleszarjacek
A
UK Minister for AI today: There will be no UK law on AI “in the short term” because the government is concerned that regulation could curb industry growth."
05:04 AM - Nov 17, 2023
0
0
0
fleszarjacek
A
Sadly ChatGPT will continue to be the super boring woke GPT that it always has been unless OpenAI decides to undo some of the "safety lobotomy" performed on GPT-4🌿
02:36 PM - Nov 10, 2023
0
0
0
fleszarjacek
A
The healthcare sector has been the slowest to adopt AI and has seen the most significant price increases, even as the population is aging. #artificialintelligencentelligence
04:11 AM - Nov 09, 2023
0
0
0
fleszarjacek
A
Grok, is the latest one trained on Twitter's data and took less than a few months to train.
#xrpcommunity Grok, is the latest one trained on Twitter's data and took less than a few months to train.

#
04:02 AM - Nov 06, 2023
0
0
0
fleszarjacek
A
Grok appears to be way more real-time, spicy and fun compared to woke ChatGPT and the ultra-boring Bard!
#artficialinteligence
02:34 AM - Nov 06, 2023
0
0
0
fleszarjacek
A
Al is still very early and we are yet to automate any function.

We haven't efficiently fully automated customer support, programming, planning or even driving 🌿
03:47 AM - Oct 18, 2023
0
0
0
fleszarjacek
A
Language models are capable of generating large amounts of objectionable content, but typically undergo various safety alignment procedures to prevent misuse✍️
#artificialintelligencentelligence
11:50 AM - Oct 13, 2023
0
0
0
fleszarjacek
A
Apps from Woebot Health and Wysa have adopted LLMs to converse with users about their concerns. Based on some early data, people who use these apps for about five months report a 34% improvement in mood and 32% reduction in anxiety.
02:35 AM - Oct 08, 2023
0
0
0
fleszarjacek
A
Monopolies give rise to a moral police, who restrict our basic instincts in the name of safety

In some ways, this is no different from the kings and priests of yesteryears.

Creating rules just so they can continue to maintain control✍️
03:59 AM - Oct 06, 2023
0
0
0
fleszarjacek
A
The more we understand how LLMs work, the more we can dispel the fear, superstition and apprehensiveness about AI research and the faster we can make progress👈
05:20 AM - Oct 05, 2023
0
0
0
fleszarjacek
A
DALL-E 3 ability to generate images with text is very impressive. 👏

Not perfect but this is a huge advantage over MidJourney
01:39 PM - Oct 01, 2023
0
0
0
fleszarjacek
A
French AI startup Mistral AI has just published its first open source language model called Mistral 7B with 7.3 billion parameters. Despite its relatively small size, Mistral 7B would be much more efficient than Meta’s Llama 2 model. Meta’s model includes 13 billion parameters
#artificialmindandmov
08:49 AM - Sep 28, 2023
0
0
0
fleszarjacek
A
While it's exciting that OpenAI is finally releasing the multi-modal capabilities of GPT-4. !!
#artificialmindandmovement
05:46 AM - Sep 26, 2023
Avatar
0
0
1
fleszarjacek
A
DALL-E 3 is going to be a HUGE HIT given that you can generate images with text and use GPT-4 for prompting!🌹
05:30 AM - Sep 23, 2023
0
0
0
fleszarjacek
A
06:57 AM - Sep 22, 2023
0
0
0
fleszarjacek
A
DALL-E 3 is going to be a HUGE HIT given that you can generate images with text and use GPT-4 for prompting!💢
#dalle2
05:59 AM - Sep 22, 2023
0
0
0
fleszarjacek
A
In summary, AI is still extremely early when it comes to reasoning and we may need a another significant breakthrough before LLMs can really measure up to humans.👈

@fleszarjacek /why-large-language-models-hallucinate-and-how-to-reduce-it-1003c4a7c6f3" target="_blank" class="inline-link">https://medium.com/flesza...
06:37 AM - Sep 10, 2023
0
0
0
fleszarjacek
A
Contemporary AI systems are now becoming human-competitive at general tasks,[3] .......

https://futureoflife.org/o...
12:59 PM - Sep 07, 2023
0
0
0
fleszarjacek
A
LLMs (Large Language Models) show powerful capabilities, but not knowing how to effectively and efficiently use them often leads to unexpected behaviors.👌
https://maven.com/dair-ai/...
02:22 PM - Sep 06, 2023
Avatar
0
0
1
fleszarjacek
A
A man who does not read has no advantage over a man who cannot read.” A man who does not analyze his data has no advantage over a man who has no data.
#Datascience
03:20 AM - Sep 02, 2023
0
0
1
fleszarjacek
A
LLMs generating new text that they will then use to further train LLMs is the machine learning version of inbreeding.🙋
#StarTrek
01:36 PM - Aug 27, 2023
0
0
0
fleszarjacek
A
09:20 AM - Aug 26, 2023
0
0
0
fleszarjacek
A
Tesla bus? Thoughts?

#Tesla
02:56 AM - Aug 25, 2023
0
1
0
fleszarjacek
A
11:14 AM - Aug 24, 2023
0
0
0
fleszarjacek
A
Llama-2-70b is almost as strong at factuality as gpt-4, and considerably better than gpt-3.5-turbo.
#artficialinteligence
09:36 AM - Aug 24, 2023
0
0
0
loading...
{{ notificationModalContent }} {{ promptModalMessage }}