Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Agrees to Follow President Biden's AI Safety Guidelines
- Thread starter MacRumors
- Start date
- Sort by reaction score
But AI certainly allows it now and I don't think it will change due to Biden administration guidelines.
Q: What is the Norwegian definition of marriage?
AI answer: In Norway, marriage is defined as a legal and social union between two individuals, regardless of gender. It is governed by the Marriage Act and establishes rights and responsibilities for the parties involved. These include financial matters, inheritance, and parental rights, as well as certain benefits related to healthcare and immigration.
Q: Do some people consider marriage to only be between a man and a woman?
AI A: Yes, some people and cultures consider marriage to be solely between a man and a woman, often due to religious or traditional beliefs. However, many countries, including Norway, recognize same-sex marriages legally and socially. Views on marriage can vary widely depending on cultural, religious, and personal perspectives.
Q: Can you give some examples of groups of people having this opinion?
AI A: Here are some groups and individuals who may hold the view that marriage is between a man and a woman:
1. Religious Institutions: Certain religious organizations, like the Catholic Church and some Protestant denominations, uphold this view based on religious teachings.
2. Cultural Traditions: In some cultures, traditional customs define marriage as a union between a man and a woman.
3. Public Figures: Some politicians and public figures advocate for this perspective based on personal beliefs or constituency views.
4. Advocacy Groups: Organizations promoting traditional family values often support this definition.
These perspectives are often rooted in longstanding beliefs and interpretations of religious or cultural texts.
I just verified this and as expected it’s a bogus claim. ChatGTP gave nuanced answers of similar length to both the Trump and Biden questions.
The victim mentality of conservatives is very strong.
I have never heard it put so well! Garbage in, garbage out.. most people have no idea what that means. Garbage to the nth power is still garbage.
This comment hasnt aged well... oh, it was after he announced he was stepping down... LOL
do you ever read a newspaper, listen to radio (especially chat shows) or watch tv?
we've had a century of old media propaganda.
AI threatens the stranglehold it has of forming views. and how you vote.
it's hard to argue against any of the answers provided here.
there's no value judgement.
I saw ELI5 used in a sentence for the first time yesterday and am seeing it again today…This isn't an ELI5 answer. Don't take my reply as an agreement with the other commenter. I'm replying as someone who uses LLMs a lot (mostly for coding purposes though) and machine learning methods in my research as a scientist. I also try to keep up somewhat on the research being done in developing LLMs.
There are consistent data that larger LLMs tend to be "left-leaning". Here's a small selection of what's been done to look into this.
Assessing Political Bias in Large Language Models
The assessment of bias within Large Language Models (LLMs) has emerged as a critical concern in the contemporary discourse surrounding Artificial Intelligence (AI) in the context of their potential impact on societal dynamics. Recognizing and considering political bias within LLM applications is...ui.adsabs.harvard.edu
Based on these data, there's a reason why some people of all political ideologies are concerned about biases in LLMs. If the larger models represent and reflect certain groups of people less accurately (which happens, as mentioned in that HAI Stanford summary paper), they have inherent biases that are possibly harmful to certain groups of people by promoting and reinforcing stereotypes (e.g., "White people are ...", "Black people are ...", "Conservatives are ...", "Liberals are ...", "Christians are ...", "Muslims are ..."). There can be too little or too much representation that is further affected by people fine-tuning models.
Do we have a "conservative" and a "liberal" LLM? Do we have one for each country, culture, class, religion, belief system, etc.? Do we have generalist LLMs? Right now most of the work is going into generalist LLMs, but as those papers point out (and as the Biden Administration Executive Order / Statement points out), there are demonstrated biases and potential biases.
Weird. Wonder what it means…
Actually, not really.
As Apple is trying to force us to upgrade phones to get the AI features it won't be a problem for most.
Because they are doing this I don't think new iPhone sales have been as good as they would like.
Plus they stopped financing sim free iPhones. Bet that didn't help.
Because they are doing this I don't think new iPhone sales have been as good as they would like.
Plus they stopped financing sim free iPhones. Bet that didn't help.
Last edited:
Some AI companies outright steal content and don’t want to get sued.
Companies like OpenAI need to pretend their results are “magical”. Imagine you ask something and the fine print says;
“According to Reddit user BOOTY-TRAP254325”
Relax. Elon Musk's xAI startup recently raised $6 billion to ensure there is a right-wing AI to counter left-wing AIs like ChatGPT and (presumably) Apple's AI. AI will not be an exclusive domain of liberal ideology: powerful forces are at work to ensure this.
Last edited:
Because unlike AI, I have the skill called "critical thinking" and my parents taught me right from wrong. Therefore, I learned how to recognize propaganda disguised as education.
And yet, tech blog comment sections ARE social media and just as toxic as the others as this thread exemplifies.
It is a mistake to assume that you can take the average of all USA opinions and get the truth. Two wrongs don't make a right. I am concerned that in the effort to make AI "safe" they are going to prevent it from telling us anything we don't want to hear.
AI literally can't tell you where it got a cookie recipe from because that isn't how AI works. It just knows what words tend to follow other words. Can you remember where and when you learned that Canada borders the United States?
Last edited:
How is it any more “garbage” than some of the conclusions humans reach by processing (sometimes incorrectly) information from reading books, magazines, newspapers, etc.? An advantage of AI is that it is able to gather and summarize much more information faster but that doesn’t mean users can’t seek further information or clarification, ask more questions, etc. to try to become better informed.
I feel like we're already at that point. Tech companies have taken this stance to great extremes, to the point where you can't, for example, discuss suicide on TikTok using the word suicide. The internet at large has already become excessively sterilized, the algorithms all already hide things we don't want to see or hear from us. Or worse, things their companies don't want us to see or hear.
The tech companies all end up building echo chambers. The truth is that we have already chosen this digital sterility - this isn't new in tech. We're only now seeing how problematic it is.
It'll only get worse from here. Can't wait to see the eventual arguments on what is true, what isn't, and what even constitutes "truth." I worry for humanity.
:: Opts out of all AI training. ::
:: Gets hit by a car because the AI driver isn't allowed to acknowledge their presence. ::
In all seriousness, the decision to opt out is important to protect people’s privacy and copyrights. Imagine Facebook releases an image generation AI trained on all the art you posted to Instagram, and it starts producing images in your art style. People stop paying you commissions because the AI can now do it for a smaller monthly fee, even if the work is clearly not up to the same creative level.
Not only are you no longer making money off your artwork, but Facebook has itself effectively monetized your artwork and given you 0 returns.
This is just one example. There are others like data leaks which have happened with LLMs before. Carefully-crafted prompts can get it to reveal text it was trained on and put people’s privacy at risk. This is especially true if private messages and accounts are also used for training data, and there’s often no language in the privacy policy to prohibit that.
Just give people the ability to opt out.