“There is a world in which generative #AI, as a powerful predictive research tool and a performer of tedious tasks, could indeed be marshalled to benefit humanity, other species and our shared home. But for that to happen, these technologies would need to be deployed inside a vastly different economic and social order than our own” - Naomi Klein
“The concern is that machine-generated content has to be balanced with a lot of human review and would overwhelm lesser-known wikis with bad content. While #AI generators are useful for writing believable, human-like text, they are also prone to including erroneous information, and even citing sources and academic papers which don’t exist.”
@ErikJonker Hmm, this is of course no unstoppable force of nature. First, we should regulate #ai and address ethical concerns. Moreover, we need to figure out which economic policies are required to ensure that this #tech improves the standard of living for all, and not just reduce labor costs and increase profits for corporations. #llm
The only #bullying allowed on #Mastodon is of #crypto and #ai. It's completely acceptable to jump in somebody's replies and be an asshole if they are talking about a technology that you heard is bad.
Would you be completely shocked to discover that the original "Q" on Reddit, which spawned Q-Anon was actually an AI / LLM?
Next imagine if such a thing was informed by a super granular and intimate understanding of our thoughts and beliefs and convictions. The AI apocalypse isn't killer robots. It's mass "collective suicide", with our human weaknesses weaponized against each other.
If I used #LLM's, I would ask them to find new lyrics to #Prince's famous song, #LittleRedCorvette majlking fun of her stupidity of wearing such a relatively unique garment to an #insurrection while condemning the participants of the #MAGA mob that stormed the #Capitol one #Jan6.
s/:Did she think she was auditioning for #Outlander in an episode set in 1789, #Paris?/s
A #LLM does not chat. It continues chat protocols with likely words.
Protocols! Including the speaker marks, which are just another type of words.
One can obviously build a chat application on top of this in the spirit of the Mechanical Turk. Just in reverse: not a human in the thing, but a thing in the intelligence. It is a generic text, not a generic speaker.
No doubt that this text continuation makes a lot of sense to us.
If you dabble in #AI and #LLM at all, please read this leaked analysis by a Google researcher.
It would indeed be wonderful if near-future, incremental AI training can be done cheaply and in open-source form, outside of the corporate IP regime. I think this may lead to the widespread use of smaller, special-purpose, open-sourced models which will indeed democratize the benefits of the technology.
The dark side is unfortunately the obfuscation of source material, and the outright theft of human creativity which will be laundered through the retraining process; we will be building open-sourced models atop stolen works.
In the chaos around #NLP, I went back and re-read the beautiful article by Lawrence Barsalou on the function of language in human cognition.
Barsalou argues that language evolved in humans to support coordinated action. Archival function of language is secondary. He highlights that #CognitiveScience#Linguistics has largely studied the secondary function and made minimal advances on the primary.
@mkhoury@shiwali@UlrikeHahn Note that in this paper something was added to the #llm : memory, planning and reflection. This is an important extension to the mere transformer component. Especially the reflection adds a kind of online feetback loop, that generates a new kind of information: abstractions. And do these „agents“ really have agency?
LLMs may be components of agents, but there’s a lot more needed.
Nice article about AI, "The reality is that no one at the beginning of the printing press had any real idea of the changes it would bring. No one at the beginning of the fossil fuel era had much of an idea of the changes it would bring." https://www.thefp.com/p/there-is-no-turning-back-on-ai #AI
Laying off reporters, but starting a 24/7 TV channel, is not a good look for WaPo. A lot of resource suck for a trickle of revenue. Hurts the core product.
"It’s increasingly looking like this may be one of the most hilariously inappropriate applications of AI that we’ve seen yet." I am riveted by the extensive documentation of how ChatGPT-powered Bing is now completely unhinged. @simon has chronicled it beautifully here: https://simonwillison.net/2023/Feb/15/bing/