Why is it that every conversation I have with a proponent of current #AI (especially #gen_ai or #LLMs ) makes me dislike the technology more and trust it less?
#gen_ai
Heute ist es soweit, ich habe das erste Mal ein großformatiges "Foto" auf der Titelseite einer größeren Zeitung (in diesem Fall der Wochenzeitung "derGrazer") gesehen, welches KI-generiert ist und auch nichts mit KI, AI oder anderen verwandten, digitalen Themen zu tun hat.
Schade. Ich bin mir sicher, das es in Graz genügend Motive und auch talentierte Fotograf:innen gibt, die schnell und zu konkurrenzfähigem Entgelt ein Bild liefern hätten können.
Mit dem hätten sich Leser:innen auch eher identifizieren können, als mit so einem generischen und sterilen Bild.
[ Contagem regressiva ]
Priorizar a qualidade dos dados e criar frameworks robustos de IA responsável são essenciais para o sucesso na era da GenAI.
Desenvolver uma força de trabalho qualificada e adaptar parcerias comerciais estratégicas também são cruciais.
Focar em resultados estratégicos de longo prazo, ao invés de apenas ganhos de produtividade imediatos, permitirá às empresas explorar o poder transformador da GenAI e redefinir seus modelos operacionais, destacando-se no mercado competitivo.
https://www.linkedin.com/posts/josetelmo_the-gen-ai-countdown-activity-7199819410453266432-ECNM
My daughter is in her final year working on her PhD and is starting to add her name to a few papers #Gen_AI . I’m a very proud Dad. #DALL·E #midjourney
“The use of AI-enabled text-to-image generators, such as Midjourney and DALL-E, raises profound questions about the purpose, meaning, and value of images generally, and the production, editing, and consumption of images in journalism specifically. This study explores how photo editors (or their equivalents) in seven countries perceive and/or use generative visual AI in their editorial operations and outlines the challenges and opportunities they see for the technology”
https://www.tandfonline.com/doi/full/10.1080/21670811.2024.2331769
Many people who commented on the race issue in Gemini's generated images, would totally ignore the intentional disinfo* and its relevance to the current events!
(*highlighted by @Mer__edith )
https://dair-community.social/@Mer__edith@mastodon.world/112039363339834529
The Future of Artificial Intelligence
https://www.youtube.com/watch?v=HAiXT1mGTXc
"In this lecture, AI expert Melanie Mitchell will demystify how current-day AI works, how “intelligent” it really is, and what our expectations—and concerns—about its near-term and long-term prospects should be."
AI and the New Theocracies
https://www.linkedin.com/pulse/ai-new-theocracies-simon-wardley-rdmwe/
"There is far too much AI doom for my liking. However, there is one issue that does concern me. It is not about machines but about people. It is almost never mentioned and mainly arises from attempts to solve the above risks. The thing that gives me concern is the rise of a new Theocracy." -- #SimonWardley
A Scientist Rebels (1947)
https://www.theatlantic.com/magazine/archive/1947/01/a-scientist-rebels/656040/
"It is perfectly clear ... that to disseminate information about a weapon in the present state of our civilization is to make it practically certain that that weapon will be used." -- #NorbertWiener
AI auditing: The Broken Bus on the Road to AI Accountability
https://arxiv.org/abs/2401.14462
"...the practical nature of the "AI audit" ecosystem is muddled and imprecise, making it difficult to work through various concepts and map out the stakeholders involved in the practice. " -- #Abeba Birhane #RyanSteed #VictorOjewale #BrianaVecchione Inioluwa#DeborahRaji
OpenAI's GPT-4 finally meets its match: Scots Gaelic smashes safety guardrails
https://www.theregister.com/2024/01/31/gpt4_gaelic_safety/
"The safety guardrails preventing OpenAI's GPT-4 from spewing harmful text can be easily bypassed by translating prompts into uncommon languages – such as Zulu, Scots Gaelic, or Hmong." -- #KatyannaQuach
How enterprises are using open source LLMs: 16 examples
https://venturebeat.com/ai/how-enterprises-are-using-open-source-llms-16-examples/
"We learned of several enterprise companies experimenting extensively with open-source LLMs, and it’s only a matter of time before they have deployed LLMs." -- #MattMarshall #VentureBeat
The horrors experienced by Meta moderators: ‘I didn’t know what humans are capable of’
"A court ruling in Spain that attributes a content moderator’s mental health problems to the job paves the way for acknowledging that fact for at least 25 other employees as well"
Prompt Engineering with Llama 2
"Prompt engineering is using natural language to produce a desired response from a large language model (LLM). This interactive guide covers prompt engineering & best practices with Llama 2." -- #facebookResearch
Coding on Copilot: 2023 Data Suggests Downward Pressure on Code Quality
https://www.gitclear.com/coding_on_copilot_data_shows_ais_downward_pressure_on_code_quality
"We examine 4 years worth of data, encompassing more than 150m changed lines of code, to determine how AI Assistants influence the quality of code being written. We find a significant uptick in churn code, and a concerning decrease in code reuse." -- #GitClear
Google cancels contract with an AI data firm that’s helped train Bard
https://www.theverge.com/2024/1/23/24048429/google-appen-cancel-contract-ai-training-bard
"Human workers at companies like Appen often handle many of the more distasteful parts of training AI and are often the lower-paid, often ignored backbone of the entire industry. " -- #theVerge
OpenAI and Google will be required to notify the government about AI models
https://mashable.com/article/openai-google-required-notify-government-ai-models-executive-order
"OpenAI, Google, and other AI companies will soon have to inform the government about developing foundation models, thanks to the Defense Production Act. " -- # CecilyMauran #Mashable
Sycophancy in Generative-AI Chatbots
https://www.nngroup.com/articles/sycophancy-generative-ai-chatbots/
"When faced with complex inquiries, language models will default to mirroring a user’s perspective or opinion, even if the behavior goes against empirical information. This type of “reward hacking” is an easy way to get a high rating on responses to a user’s prompt, but it is problematic for applications of AI that require accurate responses." -- #CalebSponheim #NNGroup
Get Ready for the Great AI Disappointment
https://www.wired.com/story/get-ready-for-the-great-ai-disappointment/
"Rose-tinted predictions for artificial intelligence’s grand achievements will be swept aside by underwhelming performance and dangerous results." -- #DARON_ACEMOGLU #wired
In New Experiment, Young Children Destroy AI at Basic Tasks
https://futurism.com/children-destroy-ai-basic-tasks
"Unlike large language and language-and-vision models, children are curious, active, self-supervised, and intrinsically motivated." -- #MAGGIE_HARRISON
Client Info
Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst