Microsoft | Principal Applied Scientist Manager and Principal Applied Scientist | Redmond, WA or Mountain View, CA; NYC possible | ONSITE, REMOTE possible | https://careers.microsoft.com
At Microsoft AI's Ad Engineering Team, we're growing rapidly and seeking talented individuals who are passionate about AI and advertising technology. My group currently has open positions for two roles:
Though the roles are based in Redmond, WA, and Mountain View, CA, candidates from NYC are also encouraged to apply as remote work is an option. The roles involve leading developments that drive innovations across our advertising technology stack, leveraging large-scale machine learning algorithms, deep learning, NLP, and computer vision to enhance both user and advertiser experience.
We offer competitive salaries (see job descriptions for ranges and geo-based adjustments) and comprehensive benefits.
If you're interested in using recent advances in cloud computing and large language models to make impactful changes in the ad industry, apply with the above links and optionally DM me on linkedin (https://bit.ly/4aYTdPx) to make sure your resume gets seen.
Microsoft is committed to empowering every person and every organization on the planet to achieve more. Join us!
Microsoft | Data Scientist II | Redmond, WA | ONSITE 50%
We're the Video Experience (VX) team, part of the Microsoft AI organization. Our goal is to make consumer video experience awesome across Microsoft products. The product includes both web and mobile camera SDKs as well as the playback experience.
We work with partners across Microsoft with a focus on a consumer grade experience.
> "Constitutional AI isn’t free energy; it’s not ethics module plugged back into the ethics module. It’s the intellectual-knowledge-of-ethics module plugged into the motivation module."
while 'what is ethical' is a broad, difficult, multifaceted question, applying the model's 'intellectual' world model (that it's built from everything it's read) to it's motivation/training reward at least doesn't seem to collapse the nuance of the question.
And for sure, if the model's 'world understanding' is limited when it comes to [constitutional principle x] that will impact/limit the extent to which it gets closer to behaving according to a nuanced understanding of [constitutional principle x].
To anyone who may be pasting code along the lines of 'convert this sql table schema into a [pydantic model|JSON Schema]' where you're pasting in the text, just ask it instead to write you a [python|go|bash|...] function that reads in a text file and 'converts an sql table schema to output x' or whatever. Related/not-related--great pandas docs replacement is another great+safe use-case.
Point is, for a meaningful subset of high-value use-cases you don't need to move your important private stuff across any trust boundaries, and it still can be pretty helpful...so just calling that out in case that's useful to anyone...
At first I was impressed by how easy it was to reach a data model with chatgpt, then I laughed as I tried to tweak it and use it. I realized it didn't really have any model concepts and was just using its various KB.
I am unsure if the so called AI can think in models but so far, not but still an impressive assisting tool if you take care of its limitations.
Another point where it lacks is in logic, my daughter has a lot of fun with the book "what is the name of this book?" but she was struggling with the "map of baal" explanation, her the answer was a certain map, yet the book had another answer, I had a third one as I interpreted a proposition. I never got an answer without a contradiction in chatgpt reasoning, and the book had been mistranslated to French so one of its propositions was changed (C, both A and B were knaves) but not the answer.
> At first I was impressed by how easy it was to reach a data model with chatgpt, then I laughed as I tried to tweak it and use it. I realized it didn't really have any model concepts and was just using its various KB.
> I am unsure if the so called AI can think in models but so far, not but still an impressive assisting tool if you take care of its limitations.
I don't know. I'm using it for exactly that ("here's a problem, come up with a data model") and it gives a great starting point.[0]
Not perfect, but after that it's easy to tweak it the old-fashioned way.
I find its data modelling capabilities (in the domain I'm using it for - API services) to be rougly on par with a mid-level developer (for a handwavy definition of "midlevel").
We have data standards and agreements with those companies, we pay them to have expectations. Even then, we're strict about what touches vendor servers and it's audited and monitored. Accounts are managed by us and tied into onboarding and offboarding. If they have a security incident, they notify, there's response and remediation.
ChatGPT seems to be used more like a fast stackoverflow, except people aren't thinking of it like a forum where others will see their question so they aren't as cautious. We're just waiting for some company's data to show up remixed into an answer for someone else and then plastered all over the internet for the infosec lulz of the week.
> We have data standards and agreements with those companies, we pay them to have expectations. Even then, we're strict about what touches vendor servers and it's audited and monitored. Accounts are managed by us and tied into onboarding and offboarding.
For every company like yours there are hundreds that don't. People use free gmail address for sensitive company stuff, paste random things in random pastebins, put their private keys in public repos, etc.
Yes, data leaks from OpenAI are bound to happen (again), and they should beef up their security practices.
But thinking people are using only ChatGPT in an insecure way vastly overestimates their security practices elsewhere.
The solution is education, not avoiding new tools.
Doesn't OpenAI explicitly say that your Q/A on the free ChatGPT are stored and sent to human reviewers to be put in their RL database? Now of course we can't be sure what google, AWS etc do with the data on disks there, but it would be a pretty big scandal if some whistleblower eventually comes out and say that google employees sit and laugh at private bucket contents on GCP or private Google Docs. So there's a difference in stated intention at least..
Who in their right mind is using free ChatGPT through that shitty no good web interface of theirs, that can barely handle two queries-and-replies before grinding down to a halt? Surely everyone is using the pay-as-you-go API keys and any one of the alternative ffrontends or integrations?
And, IIRC, pay-as-you-go API requests are explicitly not used for training data. I'm sad GPT-4 isn't there yet - except for those who won the waitlist lottery.
It's really funny to see these types of comments. I would assume a vast majority of users are using the Web interface, particularly in a corporate context where an account for the API could take ages or not be accepted.
If people were smart and performed according to best practices, articles like this one would not be necessary.
I mean, if you're using a free web interface in corporate context, you may just as well use a paid API with your personal account - either way, you're using it of your own volition, and not as approved by your employer. And getting API keys to ChatGPT equivalent (i.e. GPT-3.5) takes... a minute, maybe less.
I am honestly confused how people can use this thing with the interface OpenAI runs. The app has been near-unusable for me, for months, on every device I tried it on.
> and any one of the alternative ffrontends or integrations?
And what sort of understanding do you have with the alternative frontends/integrations about how they handle your API keys and data? This might be a better solution for a variety of reasons but it doesn't automatically mean your data is being handled any better or worse than by openai.com
I wonder what the distribution of tokens / sec at OpenAI is between the free ChatGPT, paid ChatGPT, and APIs. I’d have to think the free interface is getting slammed. Quite the scaling project, and still nowhere near peaking.
To quote a children's TV show: "Which ones of these things are not like the other ones?"
Some of those are document tools working on language / knowledge. Others are infrastructure, working on ... whatever your infra does, and your infra manages your data (knowledge).
If you read their data policies, you'll find they are not the same.
To your average user who interfaces with these figurative black boxes with a black box in their hand, how is this particular black box any different than the other black boxes that this user hands their data to every second of every day?
there are plenty of disallowed 'black boxes' within the federal sphere; chatgpt is just yet another.
to take a stab at your question, though : my cell phone doesn't learn to get better by absorbing my telecommunications; it's just used as a means to spy on my personal life by The Powers That Be. The primary purpose of my cell phone is for the conveyance of telecommunications.
chatGPT hordes data for training and self-improvement in its' current state. It's whole modus operandi involves the capture of data, rather than it being used for that tangentially. It could not meaningfully exist without training on something, and at this stage of the game it's the trend to self-train with user data.
Until that trend changes people should probably be a bit more suspect about what kind of stuff gets thrown into the training bin.
Those typically have MSAs with legalese where parties stipulate what they will and will not do and often whether or not it’s zero knowledge and often option to have your own instance encryption keys.
If people are using the free version of chatGPT then it’s unlikely there is a contract between the companies and more likely just a terms of use applied by chatGPT and ignored by the users.
I simply don't give a crap if my employer loses data. I don't care if my carelessness costs my employer a billion bucks down the line as I won't be working for them next year.
"I do not take any kind of responsibility about what I'm doing, or not doing, or thinking about doing or not doing, or thinking about whenever I should be doing or not doing, or thinking about whenever I should be thinking about doing or not doing".
As a moral questionable answering robot however, i must aks, why all things else should be tainted by the machinery, but evidence like text should not?
I am treating my employment like a corporation would. Risks I do not pay for and do not benefit from mitigating are waste that could allow me to transfer time back to my own priorities, increasing my personal "profit."
Not who you replied to, but if you agree, even a little, with the phrase, "the social contract between employees & employers is broken in the US"... well it goes both ways.
I use it because it's 10-100x more interesting, fun, and fast as a way to program, instead of me having to personally hand-craft hundreds of lines of boilerplate API interaction code every time I want to get something done.
Besides, it's not like it puts out great code (or even always working code), so I still have to read everything and debug it. And sometimes it writes code that is just fine and fit for purpose and horrendously ugly, so I still have to scrap everything and do it myself.
(And then sometimes I spend 10x as long doing that, because it turns out it's also just plain good fun to grow an aesthetic corner of the code just for the hell of it, too — as long as I don't have to.)
And even after all that extra time is factored back in: it's still way faster and more fun than the before-times. I'm actually enjoying building things again.
Pair-programming with ChatGPT is like having an idiot-savant friend who always surprises you. Doesn’t matter if the code is horrible, amazing, or something inbetween. It’s always interesting.
And I agree it’s fun. Maybe it’s the simulated social interaction without consequences. I can be completely honest with my robot friend about the shitty or awesome code and no one’s feelings are going to get hurt. ChatGPT will just keep trying to be helpful.
You can be an experienced developers with years building complex applications behind you and still find ChatGPT useful. I've found it useful for documenting individual methods or simply explaining my own/other's code or writing unit test methods or just using it to add boilerplate stuff that saves me an hour that I use elsewhere.
I think many people find ChatGPT useful specifically because they have years of experience building complex applications.
If you know exactly what you want to ask of it, and have the ability to evaluate and verify what it produces, it's incredible what you can get out of it. Sure it's nothing I couldn't have done otherwise... eventually. The productivity it enables is worth every cent.
Easily the best $20 I've spent in ages, they should have run with the initial idea of charging $42.
But holy moly anyone putting confidential information into it needs to stop
I’ve been doing this kind of thing pretty regularly for the past few weeks, even though I know how to do any of the tasks in question. It’s usually still faster, even when taking the time to anonymize the details; and I don’t paste anything I wouldn’t put on a public gist (lots of “foo, bar”, etc)
Precisely because I can abstract it is why I use ChatGPT. It can do the boring, tedious, repetitive stuff instead of me and has shown me the joy of using programming to solve ACTUAL problems yet again, instead of having to spend hours on unimportant problems like "how do I do X with library Y".
I just want to point out that some of the statements in your comment seem to be presented as facts when they are actually heavily disputed.
> "a cartel of traditional and social media companies worked with the FBI and intelligence agencies to push a false narrative (the fake Steele dossier, paid for by the Hillary Clinton campaign)"
- while there are differing opinions on the veracity of this claim, it's important to consider the potential implications if it were true. If it were true, it would suggest that the President was compromised and had a conflict of interest between his personal vs the national interest, which is a serious concern.
- Additionally, there are many other instances where the President's conflicts of interest have been on display, such as his actions in Helsinki and his use of the presidency to enrich himself and his associates.
> "and to suppress a major story about corruption by a Presidential candidate (Hunter Biden's laptop)"
- It's worth noting that this claim was heavily promoted by the Trump campaign and lacked substantial evidence.
- And even if the worst version of these claims were true, they would be minor in comparison to other serious allegations against the former President, such as the Zelensky quid pro quo and the use of taxpayer dollars to benefit his businesses.
--
(I'd also add to the broader discussion that)
- While it's true that Western media has made mistakes and increased the spread of divisive content, the narratives about China and Russia are often rooted in solid ideological (or at least mostly self-consistent) principles, such as anti-authoritarianism and free speech.
- The real issue with our media, in my opinion, is its role in amplifying domestic right vs. left divisiveness, which weakens the United States in the world, and is tearing our society apart.
The fact the FB and twitter worked with intelligence and politicians to suppress dissent is not some opinion, it's a documented confirmed fact. Denying the facts is not "opinion" , it's just lying.
Same for Hunter laptop BTW- they had the ultimate evidence, the laptop itself. FBI had it, and lied about it. That's a fact. We didn't know it's a fact back in 2020, but now we do. So let's not pretend as there are some matters of opinion here - there are facts, and there are lies.
> the narratives about China and Russia are often rooted in solid ideological (or at least mostly self-consistent) principles, such as anti-authoritarianism and free speech.
Please. It's rooted in competition and antipathy, us vs. them. Let's not kid ourselves.
The Dossier is a fake, and it was paid for by the Hillary Clinton campaign. The FBI also reached an agreement to pay Cristopher Steele $1 million to continue work on the dossier. Apparently this deal was scrapped and the payment never went through, but again, this is pretty disturbing that it went that far.
Also, the Hunter Biden laptop has been confirmed to be authentic and not Russian disinformation. The story was originally broken by the NY Post, but it was heavily suppressed by other media outlets and the NY Post’s social media accounts were blocked, as were those of pretty much anyone that attempted to share the story. The Intelligence community worked pretty much in lockstep to discredit the laptop as Russian disinformation.
(Yes, I know Wikipedia, but the only sources that really reported on it are Conservative-leaning media outlets like NY Post, Fox News, and Daily Wire, so I thought Wikipedia would be considered a bit more neutral.)
thanks for providing those sources, but I still think it's important to acknowledge that the claims you made are heavily disputed
what you stated were your conclusions (which may or may not be true):
- that the dossier was paid opposition research is not disputed, what is disputed is the veracity of its contents. Nothing in the link you shared suggests it was confirmed fake. In fact, many of the claims made in the dossier have been corroborated by subsequent investigations.
- On the other hand, while the the Hunter Biden laptop (also opposition research) has been authenticated, the veracity of the claims made about it are still in question, and many media outlets declined to run the story due to a lack of supporting evidence.
- Another difference to highlight is that Christopher Steele was an experienced professional who followed standard journalistic practices to protect his sources, etc., whereas Rudy Giuliani's handling of the Hunter Biden laptop (to put it very lightly) has been criticized as unprofessional and lacking in evidence.
- I'm not saying that bias in media coverage doesn't exist, but news organizations have a responsibility to verify stories with evidence before publishing them. It's also worth noting that the burden of proof is higher for claims that are "difficult to verify, yet easy to fake" (So it's not surprising that some stories are given more attention than others depending on the strength of the evidence supporting them)
> Nobody is going to use a chat-bot if it is (a) constantly trying to sell you something or (b) going to give you biased results that are influenced by advertising spend
I guess this is something that Google/search engines and other media like podcasts have had to navigate.
While it's not obvious how to best do it in this interface, I dunno, you still have a whole browser window to work with and the search engine solution (show the ads separately and labeled at the top) and the podcast solution (read the ads separately and clearly labeled) weren't super imaginative and seem to be working okay. Gmail ads on the right margin is maybe another example to point to.
I imagine there are probably plenty of UI things to experiment with that don't involve masking the advertisement as expert advice.
But yeah while this is all being sorted out I imagine it shrinks the monetizable surface of search a good bit.
I guess it's uncomfortable to accept that all choices you make DO have some sort of opportunity cost, and I guess I am not too surprised by the backlash against this 'college kid' on this thread. But I think the point he makes is worth raising and reflecting on (and maybe if you focus on temperance and not abstinence, it's a little easier to consider).
I think one could complain that the author doesn't credit some of the positives/educational or artistic value in video media (then again, netflix's 'hit engineering' is not necessarily optimizing for those if they don't increase addictive engagement) and that maybe he should consider reading and reflecting on things like Bertrand Russell's essay, [In Praise of Idleness](https://harpers.org/archive/1932/10/in-praise-of-idleness/)
But anyhow, after you finish the Mandalorian, reflect on how much you liked it maybe, and then reflect on what it cost you. If you aren't working on a cure for cancer or something, maybe it was in budget, but asking the question either way seems like a good practice to consider adopting
"Google's mission is to organize the world's information and make it universally accessible and useful"
(didn't dig too far into this but..) why y'all need to patent this then?
Patents can be beneficial to facilitate constructive competition, but think humanity is best served by neural nets becoming the new electricity rather than the new Apple-esque walled-garden...
I don't know what Google's reason is for this patent, but defensive patents are really common these days, and I think just about any big company has lawyers saying "patent everything you can, since you need a big patent portfolio for defensive purposes". Point is, just because Google is filing this patent doesn't mean they intend to stop others from using this approach. A link to a patent application isn't enough context to know.
If Google intends to patent defensively, they should immediately pledge on this patent. One thing I find very unfortunate, is while Google claims they're only really intended to use their software patents defensively, they've pledged not to with only a tiny number of their patent library.
One quick change in business strategy could turn Google into the world's largest patent troll.
Maybe HTML in-browser ads, but native, in-flow software ads and demand-based "rewarded" ads basically prop up the entire mobile software (especially gaming) market. There's only growth there, no slowing.
I think google invites this standard with the mission statements of "organize the world's information and make it universally accessible and useful" and "Don't be evil"
I had a discussion about this with an IBM representative 15 years ago at a symposium at Heinrich-Boell-Stiftung in Berlin: IBM's point was that one can always join their patent-pool (of defensive patents) which means you give them a free license to use your patents and you are free to use their "defensive" patents. This is completely broken: You never know which patents become relevant and any new player already lost because one just cannot keep up with a company that can extort a free license from you and then dump a few (hundred) million into development based on that.
I don't know if this is the reason, but I think Google started getting a lot more paranoid about hoarding patents when Apple and Microsoft started going after Android OEMs in the early years. Google didn't really have any "counter-offensive" patent strategy then, which is why it went on a patent buying spree back then, although most of the available ones also got bought by Apple and Microsoft through Rockstar and so on.
So best case scenario, Google doesn't want to be caught with its pants down regarding patents. Worst case, it wants to "own" deep learning, so that nobody can really compete with them. Although I think that would be a little in conflict with their strategy to open source tensorflow.
To really figure out on which side Google is now playing we'll have to see how they respond to future patent reforms, and whether they join Microsoft and IBM to once again kill those reforms, or support the reforms to abolish software patents or drastically reduce their damage.
> Worst case, it wants to "own" deep learning, so that nobody can really compete with them. Although I think that would be a little in conflict with their strategy to open source tensorflow.
Only a little. Releasing all the models and frameworks helps advance the field, helps with finding people to recruit, helps with integrating them into teams, and so on. This is why so many giants find it in their own self-interest to contribute to FLOSS these days.
Competition-wise, as is often said, Google has all the data. If for every deep learning advance they make $1 and the competitors make $0.95, they win. Patents here are quite helpful: you may make a neat translation app using some new tricks, and then discover when you go to commercialize it that oops, Google's patented 'using neural nets for translation'. Then you either quit, get sued, get bought, or give them most of your profits.
I rather no company patent them than any do-no-evil company. Besides why do you think they won't use/abuse it? The whole point of spending so much ridiculous amount of money on a patent is to get exclusive rights to be the only one to use it. Isn't competition better than stagnation by a single company?
Nope, but that doesn't matter - as enforcement is the concern, and there prior art shines. If the motive is purely defensive (and I don't include the menacing of a portfolio in that category), then it is the ideal move to make - as bad actors waste more energy filing poorly researched and easily refuted patents.
Probably doesn't prevent the granting of a patent, however it makes it impossible to enforce the patent. Mike and I published this in 1998 http://www.rage.net/wireless/wireless-howto.html . A Cisco legal team found this in 2008 and contacted me because the owners of patent #7035281 were coming after them. Doing a simple write-up of what I thought was obvious at the time - stick a wireless card into a Linux PC and have it route packets - may have saved all of us from having the wifi router in everyone's home restricted by patents. So whatever ideas you implement, be sure to blog about them and make sure archive.org gets a copy.
The problem is that other people will just create one or more patents which are around using your breakthrough algorithm in different contexts. The famous amazon one-click shopping patent as an example. The internet is the breakthrough, but because there's no patent on that, its easy to surround with patents that should be too obvious to be patents, but have legal teeth even so.
If you patent the core idea, the other patents become a lot less useful. (Not that I think Google is thinking this way. It's just a PR problem to them. When no one is looking, I bet they do whatever they can to get as much money/power as they can.)
It is a terrible situation, and putting out prior art certainly doesn't fix that - but the alternatives courses of action are worse (assuming purely defensive interests). I think you're right though, if the internet was somehow patented then we wouldn't have one-click shopping patents... or much of anything really - we'd likely be having this discussion over a Minitel service.
Consider the development of Microsoft's attitude regarding software patents. There's no reason whatsoever to think these patents are in good hands with Google, especially considering that their Android licensing policies seem to be carbon copies of old Microsoft tactics.
Their goal as it seems up to this point is to get a monopoly on the framework and technology with opensourcing of their framework. Good for hires and keeping the competition in check. A patent will tighten their grip on ai.
Unfortunately, someone else can come along and patent a trivial, supposedly non-obvious "next step" invention, and the original inventor would have to pay to license the slightly improved version to remain competitive. With a patent on the main technology, they could probably negotiate a reciprocal deal at the very least.
What are rules for such a publication? I guess that something published in IEEE shouldn't be patentable and have my doubts about publishing on a blog, self-hosted, in Romanian. Where is the line according to the US law?
To be sure, you submit the provisional patent application, and never file the utility. It's only $260 to file. After 1 year the provisional dies and you have a permanent record of the prior art.
In this case there is a provisional patent from 2014, and this application follows from that provisional.
I've been reading more about this. Public disclosure vs. provisional.
In the US you get a year after public disclosure to file even a provisional. But not so if you want non-US patents. So you are closing some doors but keeping others open with public disclosure. Also, you are starting a 1 year clock.
A provisional is private; it does not count as public disclosure. A provisional is nothing more than a priority date, assuming what you have disclosed in the provisional itself is sufficient, and novel. You can even refile the same provisional every year as long as you believe the subject mater is still novel, but you get a new filing/priority date each time.
As for preventing a 3rd party from patenting the subject matter, either one is sufficient. However, if you publically disclose, only then you also get protection from a 3rd party who builds on your work. So in that case public disclosure is better than the 'secret' provisional.
Better yet, just timestamp the document in the blockchain.
Chances are that before you get your thing published on IEEE, someone else will start and finish writing an application and maybe even get the patent granted...
I'm familiar with the implications of quantum computing on factorization and thus DEcryption but have heard very little about about quantum computing enabling ENcryption until the last paragraph of this article
("...Chuang expects to see quantum encryption methods that will inscribe sensitive data into the very states of atoms")
Very curious about current state of this research (relative to the current state of quantum decryption)--any experts in the room?
Not an expert on Quantum cryptography, but an ex particle physicist here. The basic idea is that using a quantum channel (sadly means new hardware, so not over tcp), eavesdropping becomes impossible without destroying the quantum state of the signal (guaranteed by the laws of Quantum Mechanics). If an eavesdropper intercepted a message, that would be detectable and you can drop that packet. Wikipedia has a good intro: https://en.wikipedia.org/wiki/Quantum_cryptography
Theoretically, if you intercept in the middle, you destroy the pattern that you observe. This is a physical quantum effect, and will happen no matter what hardware you use
Since the intended use is key distribution, a MITM is fine as long as you can detect it reliably: you can keep sending new keys until one isn't eavesdropped upon, and then use that key.
I'm not talking about eavesdropping, I'm talking full on MITM. Cut the connection and insert a middle man. Both sides think they're communicating with their intended target, but they're communicating with you. How does quantum crypto protect you from that?
If someone intercepts the quantum key, it will modify it 25% of the time. If you randomly measure (and verify publicly with the sender) a fraction of your total key and find it unmodified, it means the rest of the key probably is too, up to a certain security factor. By starting with a longer key and measuring more of it (or doing privacy amplification, for example xor-ing multiple keys together), you can get as much security as you want. It also means the security is everlasting, meaning someone cannot retroactively break your key in 100 years using some mega-computer.
Maybe it was for a particular implementation? Funny story: the first toy impletementation of Quantum Key Distribution used a device with rotating photon polarizers. Quantum Key Distribution is completely secure so on paper the device was too. However, you could actually hear the polarizers rotating in a way you could intercept the whole secret key... as long as you were not deaf!
Cryptographers are generally not interested in using quantum physics to achieve secure communications
Cryptographers interested in encryption schemes that use mathematical structures that are not amenable to any known quantum algorithm. Lattices, Ring Learning With Error
Cryptographers are also interested in how Quantum Computers will scale to large sizes. It will be important to understand what the largest quantum computers that can practically be built are.
Quantum cryptography is still applied in some places. Switzerland is using quantum crypto below lake Geneva using optic fiber. Recently, China announced a quantum crypto satellite program.
This has never made any sense to me. What is the threat model? In almost any case where you can use QKD, it would be cheaper to send a courier with locked/handcuffed brief case.
Nearly 40 years old, we actually have an algorithm that is not known to be vulnerable to quantum attacks: https://en.wikipedia.org/wiki/McEliece_cryptosystem Quantum computing is nothing to worry about. (Though I admit it is interesting to think about what sorts of encryption/decryption schemes are only feasible with a quantum computer.)
The real problem is that quantum computers would break the forward secrecy algorithms we're using today. So if a practical quantum computer shows up tomorrow then we can start using different cryptography tomorrow, but evildoers will be able to decrypt everything we're sending today just by storing it and waiting for quantum computers.
Which is a pretty big "oh crap" that will catch a lot of people by surprise if quantum computers ever really happen.
Here's food for thought: What would be required to make a nontrivial fraction of TLS traffic on the web post-quantum secure? How much would it cost?
What about other legacy systems?
Actual, physical quantum computing is still in its infancy, so sure, it's hard to conjure up the wherewithal to worry. But nonetheless we have a long way to go before we can say the world is ready for real quantum attacks.
Quantum crypto is a large field. One aspect is Quantum Key Distribution (QKD). It opened the door to the whole field of quantum computing (it was discovered before Shor's algorithm).
QKD allows you to distribute a one-time pad while only sharing an authentification key. It is (on paper, if you don't count experimental flaws) theoretically-secure, meaning you can't break it or man-in-the-middle attack it even if you have infinite computational power (with 1-epsilon probability, epsilon being as close to 0 as we decide). In practice, most QKD systems can be hacked through hardware flaws.
At Microsoft AI's Ad Engineering Team, we're growing rapidly and seeking talented individuals who are passionate about AI and advertising technology. My group currently has open positions for two roles:
1. *Principal Applied Scientist Manager* [Apply here](https://jobs.careers.microsoft.com/global/en/job/1681254/)
2. *Principal Applied Scientist* [Apply here](https://jobs.careers.microsoft.com/global/en/job/1696716/)
Though the roles are based in Redmond, WA, and Mountain View, CA, candidates from NYC are also encouraged to apply as remote work is an option. The roles involve leading developments that drive innovations across our advertising technology stack, leveraging large-scale machine learning algorithms, deep learning, NLP, and computer vision to enhance both user and advertiser experience.
We offer competitive salaries (see job descriptions for ranges and geo-based adjustments) and comprehensive benefits.
If you're interested in using recent advances in cloud computing and large language models to make impactful changes in the ad industry, apply with the above links and optionally DM me on linkedin (https://bit.ly/4aYTdPx) to make sure your resume gets seen.
Microsoft is committed to empowering every person and every organization on the planet to achieve more. Join us!
Applications close on May 14, 2024.