3,508 karma · joined December 21, 2018
But the public reception doesn’t really tell us much about the strength of the conclusions. For which there is overwhelming consensus, very clear mechanisms of action, and pretty straightforward directional signal.
Are you questioning the comms or the consensus? I’m not clear from your response.
I disagree, this deserves continued conversation and your comment is inappropriately dismissive. The evidence that companies are secretly using mics from connected devices from major manufacturers is scant; on the other hand, the technical evidence and forensic analysis is massive this doesn't occur.
My take, this looks concerning due to their intent but I don't feel it's technically possible given the other analyses that have been done. MindSift states their goal, but it's not worded like this is in prod or how it would work. (Perhaps they intend to use UI tricks to enable the mic; very different than doing it covertly.)
As for CMG, they point to their partners but do not say those partners have any hand in this work - they could be using a "mic on" trick to programmatically change ad targeting in those networks, which would be less concerning than if those giants were doing this.
TBH, if any of the tech giants are doing this, it'd be a pretty major violation of various laws/FTC agreements, etc. If it were happening, someone should whistleblow pretty soon and be in line for a major payday - this violation would make the $5B from Cambridge Analytics look small.
Yes, I wanted to work for them so badly it was well worth the risk. Sometimes you see opportunities and want to pay for them.
This can be challenging in the decision making sphere, or when events are unpredictable (planning, stock trading). But it would be weird if we didn't allow for deference to people with actual specific knowledge and experience over those who are just surveying from the outside.
Even further, maybe what you're saying is true under general assumptions, for 90% of companies. However, how do you know there aren't outliers for whom the assumptions do not hold true?
Not to be too blunt, but what evidence or information do you have that makes this more than an "armchair quarterback" situation? Have you experience in this particular field to such an extent that it can be extended as a functional rule within the industry, or are you expecting it to behave like physics? We can't just take general principles and assume they hold true in specific circumstances - that is what makes this kind of work so hard.
It feels wrong to say “this approach won’t work anywhere”, maybe you are saying this approach won’t work everywhere. That’s fine, it doesn’t need to. And I’d defer to the person who did this work, it sounds like successfully, versus someone whose opinion is very broad and I’m not sure what evidence or experience they are basing their commentary on.
It’s one reason Mac Apps get bundled as a single “file” from the user perspective. You don’t have to “install”, you just copy one file with everything. It’s a simpler dev experience.
Sure there are tradeoffs, but that’s great! We should accept that tradeoffs mean people can chose what works best for their specific context, rather than “best practices” which are silly.
In the 1970s, the USSR would literally just buy the latest US chips. No need for much crazy subterfuge and honeypots - go to conference, find someone with chips who likes cash, and buy them. That's what led to the crypto bans in the 1990s, where software that implemented cryptographic technology couldn't be exported - as if bits cared much about crossing borders and clickwraps!
Or just stop responding to nonviolent crime, which already happens. We’ve got to be realistic about the tradeoffs.
I don’t see how they shouldn’t be able to take less money themselves as long as the other creditors also got more money.
What experiences/evidence do you have that informed your opinion? It sounds like you've had pretty negative experiences.
I wonder if the "optimal" theory is portfolio in this case, or if there is a new generation of VC/pharma investors who want a higher probability at a lower return.
The ability to constantly change and backtest as your audience changes (because source traffic is a dynamic block more akin to a river than something static that we can test once for!) always sounded powerful.
I don’t think this is accurate, wire fraud is literally named after reaching across jurisdictions to commit crimes via telegraph wire. What is t resolved is when there is no escalation path.
In these situations, I’ve been able to sufficiently program the agent that I haven’t seen too much of an issue as you described. Consistency is a feature.
I, personally, script meetings because I don’t communicate effectively when put on the spot. To me, “unscripted” interviews feel like gotcha journalism - why not let someone take their time to describe their thought process?
POTUS is not in a tactical fighter jet dogfight where low latency is massively important. What matters is thinking things through, to consider second order effects.
Instinctual reactions are great for sports but I have no idea why they’d be important for politics, outside of optics. Type II systems, not type I.
Alternatively, if everyone is clamoring for the drug, maybe that influences regulators/lawmakers for what gets covered by Medicare. Honestly, the way people are talking about semaglutide - it could become a drug that is just provided to everyone who wants it, by default, paid by the government. (And would probably be a net positive ROI by the government - the amount of savings and increased productivity if America reduced obesity rates by 10+% would be incredible)