I'm choosing to read this as an option for people who want to put forward a serious attestation that the writing is not generated by AI, and are looking for ways to make that pledge tangible and give it a level of authority that feels more significant than "I promise" by getting that event-stream sent and signed. In that light, it seems like a useful offering!
I think it's important to consider the risks/rewards/benefits. There's definitely a sense in my circle of contacts that if a work is seen as purely human then it's somehow better and more authentic, and annecdotally, it's also possible to win points by taking something created with the help of AI and passing it off as your own independent work. Like social credit, there's a sense that you'll seem smarter than you feel yourself to be.
With that in mind, absolutely any technical solution to detect AI or attest to human authorship will be abused. The only context that a solution like this one supports is one where there isn't a risk or reward, the author just sincerely wants you to know that it's human-produced.
Plenty of people, after all, produce a draft with AI, then type it all out again editing and refining and updating as they go, and then ask an AI to look at the result and make suggestions, and then go back and make the changes they agree with. Such a workflow would be deemed human by semoi - so it's lucky that there's no point in lying about it.
I’ve been wanting to get better acquainted with local inference but I don’t have the hardware, which has made me think about something I haven’t seen discussed, which is local collaboratives. The economics makes it seem like a group of people joining together to run good hardware and an open model might make sense, but I haven’t seen anything like this mentioned. Have I been missing it?
I think it would be pretty neat to launch a service helping people who wanted to participate in something like that locate one another.
I appreciate this distinction. The are multiple senses of SOTA and one that has been taking on greater mindshare is as a synonym of “the best available”. By rebasing on SOTA as generally available and understood versus cutting edge, which has limited distribution and leads the way, we expand the vocabulary we have available to describe what’s going on. Thanks.
That’s a good point. So maybe another point of divergence here is that the outcomes of the DMA are rooted in inherently unpredictable market interactions, whereas a medical device standard depends on the device performance and characteristics.
I don’t think it makes sense to create an accountability framework for a company that requires the cooperation of the market, because I think companies should be in a position to either comply or be held accountable on their own merits
It sounds like what you’re saying is that because the legislature can’t anticipate how companies will abuse loopholes, they sidestep that by outlining the outcomes instead.
The issue I have with that approach is that I don’t agree with that approach to governance. I believe it’s incumbent on the regulator to define what is acceptable vs. disallowed in unambiguous terms.
Big scary warnings aren’t a solution either. I watch the way my son interacts with consent screens and warnings, and it’s barely believable- the average person is very well trained to click through the warnings.
Someone might believe that people who ignore the warnings deserve everything they get, but I respectfully disagree. I remember helping my grandma uninstall and remove all the hostile browser extensions that had tricked her into installing them. If Apple is protecting vulnerable populations by taking the choice out of the users hands, even if it’s only profit motivated, I’m okay with that until someone presents an alternative that actually addresses those needs.
Apple wants to implement features that access data locally. It doesn’t want to allow competition for offering those features, but if it did, competitors may use that access to local data to exfiltrate.
So it is about both competition and, as a result of creating competition, privacy.
As I follow the situation, it seems that regulatory uncertainty is a major issue though- the EU’s requirements are framed in terms of outcomes sought, rather than in terms that can be quantitatively shown as met or broken. So it’s not a matter of dedicating a team to meet a list of requirements, but instead navigating the worst case scenario of enforcement if post-launch the EU determines that the proscribed outcomes aren’t being met.
There is a vast asymmetry in knowledge and capability to make those choices for most users. Most users will press agree and consent to things they don’t understand to get to the next screen, and while that would be terrible for the individuals I think it’s also important to look further at the ramifications because probably Apple would be blamed.
I like it I think it’s sort of cool to see the different environments around Apple Park and be able to hear from a lot of different employees without having to watch a parade over the stage
Plenty of comments about using an LLM to assist with this, and I was happy to be able to read about a learning experience where the stakes were pretty low and the feedback loop pretty tight. Thanks for writing it up; for me, it reminds me that some of the use cases where an LLM might be an efficient tool are also the places where it can be wise to take the opportunity to learn and sharpen new skills.
For an adobe creative cloud subscription? I just looked at their pricing page and could only locate "annual, billed monthly" options. If there's an actual monthly subscription which is just a little more expensive then I'd say the insidious aspect is in hiding it.
If I read it right, the 4B ballpark figure is based on total annual per employee cost of 500k * 8000 employees, so the figure is actually 4B/year. 20B over five years.