Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

cisa.gov

Post reply on HN
Print view View original

161–170 of 195 posts

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#161

Post by broscillator »

superkuh wrote: post #125
lcnPylGDnU4H9OF wrote: post #108

Earlier quoted context omitted.

The difference is that rather than altered, they’re generated. It’s a new seeming-photograph that can’t be discovered as an alteration by checking against other photos to play “spot the difference”. Nobody really had this capability before.

>Nobody really had this capability before. I find your distinction between altering and creating to be ... a sidetrack. But okay, let's pretend I said "Deepfakes is just the hype word for digitally altering or creating photos/videos". Large institutions/etc have definitely been able to create novel "photos" from scratch for a long time now. It was just a lot of work, which is why it was restricted to large groups of…

I don't think photos are the real issue, it's video with voice.

Think of telephone scams, and now apply deepfake versions of a family member to it. Imagine your 70yo aunt having to deal with knowing if the whatsapp audio or zoom call that sounds/looks like you is really you.

I don't think this was possible before, feel free to prove me wrong with some evidence.

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#162

Post by ziddoap »

superkuh wrote: post #93

There are no deepfake threats. Deepfakes is just the hype word for digitally altering photos/videos. In the past doing this well was restricted to governments, large corporations, large institutions. And they did it regularly. Now individual human people can do it themselves. That's good. That's not bad in any form or manner.

>That's not bad in any form or manner.

I can think of several ways that these types of technologies being easily accessible to anyone who wants it directly leads to bad things.

There's good things, too. But come on... Not bad in any form or manner?

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#163

Post by wutwutwat »

squeaky-clean wrote: post #135
DharmaPolice wrote: post #95

Earlier quoted context omitted.

I must admit I found that element of the story...surprising. Has realtime faking gotten that good yet? Presumably there was back and forth in this call so this person was either disguising their voice or typing responses to be generated on the fly. I know all this can be done, I'm just surprised it's reached the maturity where an attacker would choose to impersonate someone the call recipient presumably knew vs just…

You could probably reduce the "delay" by using a soundboard of pre-generated filler material and playing that while you type the real response. "Let me find that bookmark", "So the thing about that is...", "ummm yeah. so...", "hmmm no not really" You can also use text macros to type the response faster. Here they were trying to get MFA access, so you could map longer phrases that will come up often like "Okta multi f…

Why type or use a sound board. You aren’t thinking mission impossible enough

Live transcribing in realtime has been a thing for, forever, so there’s no reason for me to think this couldn’t all be glued together into a “voice changer” like the typical super deep “I have your son give me a million dollars” boxes, except instead of doing frequency modulation it is pipes to a model trained on someone’s voice, and applies it. Transcribing to text probably isn’t even needed because why would it be for machine to machine modification. It only needs to go to text for human consumption.

Raw pcm bits from audio in -> AI model trained on victims voice -> line out to phone or voip app.

We totally have the compute to do that. Probably with our phones.

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#164

Post by iudqnolq »

mistrial9 wrote: post #137
lobf wrote: post #120

Earlier quoted context omitted.

Please link me a single story you think NPR was untruthful about.

NPR dutifully ran the "Saddam Weapons of Mass Destruction" *fake* 100s of times with utter sincerity. ps- also the "dumping babies on the ground" fake too IIR, but maybe only a few times on that one. Congressional testimony on camera by a daughter of a State Department official IIR to justify the Kuwait invasion by Bush I

True, but not evidence of the left wing bias originally claimed at the top of this thread

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#165

Post by sumthingsumthng »

lobf wrote: post #120
DrThunder wrote: post #103

Earlier quoted context omitted.

Reliably left wing? Yes. Reliably truthful? No.

Please link me a single story you think NPR was untruthful about.

not really possible since the people that planned and staged the thing the story was about were pretending to actually do something when they hired people and rented things to fake an origin story that was then covered by the news while another team that the first team wasn't aware of went a bit overboard and brought in surface-to-air missiles and shot down a passenger plane full of dying people.

you see, the issue is not NPR being untruthful. the issue is NPR not even having a chance to check. member how "logic"/"rationality" works on faked premises? so smooth and yummy.

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#166

Post by vorpalhex »

rtkwe wrote: post #115
perhonen wrote: post #101

Earlier quoted context omitted.

perhaps because the threat of "comply or there will be consequences" worked.

Except most of the posts pointed out to places like Twitter stayed up with no moderation applied and the worst thing that's happened to Twitter recently is it's acquisition by Musk.

Coercion doesn't have to be effective to be illegal, just like a bribe doesn't have to be accepted to be a bribe.

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#167

Post by trust_bt_verify »

jtbayly wrote: post #84
trust_bt_verify wrote: post #76

Earlier quoted context omitted.

“ A federal district court in California dismissed the claims, and the U.S. Court of Appeals for the 9th Circuit upheld that decision. The court of appeals ruled that although “it is possible to draw a causal line from the OEC’s flagging of the November 12th post to O’Handley’s suspension,” there was no “state action” for O’Handley to challenge under the First Amendment. California certainly exercised governmental au…

You didn't do a very good job trusting or verifying: https://arstechnica.com/tech-policy/2023/09/supreme-court-co... > The 5th Circuit appeals court saw things differently, finding that Biden administration "officials made express threats and, at the very least, leaned into the inherent authority of the President's office. The officials made inflammatory accusations, such as saying that the platforms were 'poison[ing…

“Be kind. Don't be snarky… Edit out swipes”

https://news.ycombinator.com/newsguidelines.html

Check them out when you have a chance. I _trusted_ ScotusBlog and the previous comments gave no references to _verify_.

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#169

Post by AlecSchueler »

mcpackieh wrote: post #158
AlecSchueler wrote: post #94

Earlier quoted context omitted.

I've seen pretty good depth estimation algorithms that work off of 2D images, even without the use of AI.

Those depth estimation algorithms can't be used to distinguish a photo of a photo from just a photo. They will report false depth in a photograph of a flat photograph.

Yes, that's my point. You can't rely on the depth map in the image metadata to be the differentiator because it can easily be faked with depth estimation.

Re: NSA, FBI, and CISA Release Cybersecurity Information Sheet on Deepfake Threats

#170

Post by cthalupa »

efitz wrote: post #146

Courts already have processes in place to challenge evidence; I’m not worried about them. The problem with deepfakes is the instant amplification they’ll get from media and well-intentioned responses from people of good faith. So there will be the deepfake of (insert your favorite politician) saying something unsavory and it’ll be taken at face value and a bunch of people will cancel their social media and bank accou…

> What if legislators were required to wait one year before passing laws in response to a public event of some sort?

Someone discovers a way to bypass regulations and legislature and easily acquire the material needed to create a dirty bomb and sets one off. Now we wait 365 days to fix the loophole.

Pretty obviously an absurd extremity for that example, but I sure as shit don't want there to be that level of lag time on important things like that, and I imagine most people would also want to be able to move faster in situations where it is dangerous not to. How do we determine where that line is, what the exceptions are, etc.? What happens when we need an exception and didn't realize it?

Post reply on HN