Make a video on how this impacts folks in the IT/cyber security careers from either a positive or negative side. Maybe talk about some roles that may be created from these problems, etc. Thanks for a great video.
Did you not watch the video? Attacks will be more sophisticated, requiring more security posture. Attacks will scale in complexity and volume as the barrier to entry gets easier with AI and generative malware. Cloud, Cryptography, AI related jobs will continue to outpace the number of people to fill them. Governance and Risk as regulators panic once the realization starts to set in.
@@Chr1s-fm6bi That will also require people to be open to the idea of adapting to the new current wave of threats and new attack surfaces. Thus, requiring further education and awareness. I am all open for that since i am just starting my career in the field.
An anecdote on Ai-written malware: very few instances have been found in the wild. The ones that have been found stood out because they were incredibly crearly written and well annotated. Hackers will tipically try to obfuscate their code, but the AI bot just can't help being a helpful little engineer.
2025 - I think AI will be a big problem, but infrastructure attacks, some of which have already happened will continue to be a massive risk. As much as the fancy stuff is scary, many industries are woefully behind on doing the simple things (stay current on patches, look for dumb exfiltration, etc). You don't need an AI attack if your software is 5 years out of date and nobody is watching your logs.
Thanks for the information and analysis. It would be useful to have reports and data sources to support how some of the these predictions grew or increased. I think another point to mention is the regulations and controls for AI. It should be the first one from my point of view. Regards
Wow! I love this person! Happy New Year 2025! To anyone reading this, I want to tell you that this year will be your year, and you will succeed in anything you're working on or already doing! Trust the process, don't forget to take care of your health, and strive to be the best at everything you do in life!
There should be a mechanism to identify if a content is human-generated or AI-generated. (Just like how youtube has made an option and asks creators to mention "Paid Promotions" if it is actually a paid video)
Thanks for yet another insightful video, Jeff! It seems increasingly likely that we may see greater adoption of Zero-Trust-as-a-Service platforms, potentially simplifying the path for small and medium-sized business to implement zero-trust frameworks without the need for building them from the ground up. This evolution could enable smaller organizations to strengthen their defenses against the rising tide of sophisticated cyberattacks.
You are good. I thought of all these points through the year. Your crisp elaboration on art of possibility, makes me worry about the amount of investment Information security departments would need to make and how fast on priority Potentially make or break
Many thanks for your video. It's insightful and succinct. Some thoughts: 1) Do you think AI could be used to scan deepfake videos in realtime by scanning for artifacts using graphics algorithms? Could we use something like triangulation (video from different angles) to prove a deep fake. Artifacts would become prevalent if it has to sync different outputs. But, deep fakes algos are improving also. 2) I worry that AI could simulate a hacker mindset and start thinking to look for vulnerbilities dynamically and exploit. This however, could be used by defensive software to identify attacking patterns.
I’m glad you liked it. On #1, I think that is bound to fail ultimately. I have video on deepfakes on the IBM Technology channel so take a look for that and you’ll see why. On #2 I think that definitely will happen, and may be happening already
As the bad things happen and the society transform, I expect to see people thinking more and lean towards living responsibly. Topics like human values will be on the table more often.
Great question. In fact, I recorded a video last month on exactly this topic. I don’t think it will post to the channel until February according to the current schedule but subscribe and be on the lookout for that. The short answer is that I think it will eliminate some jobs while creating others. AI, used correctly, is a tool that will enhance human capabilities but the bad guys will use it too so we will still need people to run the defense
Thanks for the great video I have a problem and the problem is that idk where I should start because this topic is a combination of both AI and cybersecurity and I need a clear path that will include both AI and cyber security maybe you say both of them has a clear path and I agree with that but one the example problem is: is it neccery to learn the network infrastructure for this topic? Which is necessary for cyber security
There are a number of places you can start, actually. One suggestion would be to check out the cybersecurity architecture series I did for the IBM Technology channel
Thanks for making those videos. One point though, QC is only going to break some cryptography relating to key exchange protocols, and it seems most libraries are already safe. Also just like 2024, absence of MFA, weak passwords, misconfigurations...etc are going to still be on top of daily news!
Quantum systems should be able to break symmetric crypto with Grover’s algorithm and asymmetric with Shor’s. It’s easier to address the symmetric break by just doubling the key size but we need quantum safe algorithms to replace today’s asymmetric ones
I appreciate your video presentations... very well presented and informative. I work in the film production industry and watching your videos brings to mind movies like The Matrix, iRobot and The Terminator. They used to be amusing science fiction but more and more, they sound like prophecy. I'm not an IT guy. But of course, I use computers everyday in my work. I was attracted to your video about the Dark Web because after a massive security breach a few years ago, I received free credit monitoring. Since then, I've been notified that my personal information, including my SSA, have been found on the dark web. I had never heard of TOR but a web browser I use offers browsing with TOR as an option. My question: with criminals and AI getting more and more sophisticated, and so many voices out there saying this or that (how can we even know if they're not bad guys), short of moving to a cabin in the wilderness, what are the best practices for the average person to guard against the ever increasing intrusion of our privacy and cyber security?
Thanks for watching! I’m glad you liked these videos. The best defense is education, which you are doing. It all starts with awareness and that’s what we are trying to promote on the IBM Technology channel with these videos. I’ve got a bunch more that I hope will help. More to come…
Thanks Jeff and team for these insights. Jeff can you recommend some good books for Security? I've already watched your cybersecurity playlist and want to know more elaborately. Regards, Ben
Great video and overview of things to come. These videos provide great discussion ideas for our classes that get the students engaged (don’t worry, I cite you guys as the source).
Thanks for sharing. Same thoughts I had for the new year basically Quantum and AI. I’d like to add one more on the AI related attacks, that is the Data loss/leakage by insider threat due to the use of AI by organizations. If organizations do not classify and protect their data according to their sensitivities, AI can be used by insider threats to get data that they ordinarily do not have access to.
There are trends in cyber security that point to single platform (which many would say it's alreayd a reality) alongside with challenges to build more effective tools for Cloud / Infrastructure. These are not so eye catching as preditictions for AI in cyber. But out of all my predictions, I'd stick with AI Agents and how these will impact Cyber professionals and the insdustry.... it's getting real fast, really quickly! 🏎🏎
Going into 2025 and beyond, hopefully micro-kernel operating systems will gain traction, and allow a separation of responsibilities for each operating system task. Having a monolithic kernel is a difficult attack surface to safeguard, because it's hard to reason about how one change could propagate to many places in the code itself.
Different LLMs will give different answers. In fact, sometimes the answers will vary from day to day. Some are better than others but there are many variables that determine outcomes
jeff, what percentage of how LLMs are actually functioning under the hood in praxis is still a black box? would it stand to reason that as the complexity and capability of such models increases that so to would their propensity to hallucinate, perhaps in even more alarming ways? thank you
It's certainly possible. However, the more popular models are open sourced so that they can be examined (to the extent that we have time to do so). But, verification and human in the loop (HITL) are still very important
@@jeffcrume thanks for the reply. do you have any idea as to the first part of my query? (roughly how much do we actually understand about the way LLMs are working after they are deployed; i dont mean to conflate this with our understanding of how to write the code for them to work, but rather, how that code is actually achieving the end result after the model is up and running). hopefully i am making sense. also, is IBM planning more layoffs this year?
There is no single passkey company but if the web server you are logging into with a passkey gets compromised, then all that is revealed is your PUBLIC key, which is not sensitive and can be shared PUBLICLY anyway
The shortest answer I can give is that we have a high confidence because the US NIST agency which brought us AES, which has withstood the test of time, has used the same process in selecting quantum safe algorithms. The process is open so all the world's cryptographers have an opportunity to review and poke holes
Very true. I use GenAI more as a tool to suggest to me things I already know but couldn't think of or as an intro that I then go verify with other sources
I use AI extensively in my bug bounty program and it has helped me immensely identifying, proving and reporting vulnerabilities hardening the clients Cybersecurity postures as a result.
Sounds good in principle but guess who won't put that in their version? The bad actors. So it won't ultimately prove anything as there will still be doubt
another potential for 2025 is the impact of quantum computing to crypto currencies. it might break the security of block chain were all the crypto currencies rely on. if that happens, the value of crypto currencies like bitcoin will be in trouble.
When the infosec budget is limited, use AI tools as red team to attack the organisation (pentest systems, intel gathering, phishing) and then to advise on necessary defences (patching, security gaps, training). I wonder how many would do it and how many would hope they are not the targets?
With deepfakes getting insanely good, face id authentications will start to be less secure? Today almost all my usual financial apps ask me to choose face id over touch, one time codes, etc to approve transactions
I’m hoping that won’t be the case because FaceID relies on more than just the flat image but also looks at contours and other indicators to verify that what it is seeing is a real person and not a photo or a video
I asked AI to comment on @Mrivlian comment: Here is AI answer: 😅 The comment raises an interesting point about the capabilities of AI in the context of cybersecurity and malware development. While AI can indeed assist in writing code, including obfuscated or malicious code, the effectiveness and intent behind such code would largely depend on the instructions given to it. The anecdote about AI-written malware highlights the contrast between human hackers who often use obfuscation techniques to conceal their code and AI's tendency to produce well-structured and readable code. This could make AI-generated malware easier to detect, as security systems are usually designed to identify patterns and anomalies in code. However, as AI technology evolves, it's possible that more sophisticated models could be trained to produce more complex and less detectable code. The ethical implications of AI's potential use in cybersecurity, both for defense and offense, are significant and warrant ongoing discussion. Overall, the comment opens up a broader conversation about the role of AI in cyber threats and defenses.
Phone based AI hallucinating, whilst responding to a nonexistent threat and shutting down the devices’ functionality rendering it useless. Like having a self initiated ransomware attack, committed by the AI supposedly protecting the device.
I see medical advancements happening that would help with safer surgeries and improved early disease detection. Blockchain advancements within cybersecurity and Ai. Those are some of the predictions that I thought of that did not include your own.
Whether to believe this 'UA-cam prophet' or not, one thing is certain-if cybersecurity were a religion, you'd already have us all converted! 🔒✨ Preach on, oh wise predictor of cybersecurity trends! 📈.
My prediction is: Common people will be addicted to AI like rats on cocaine. Some legislation will be enforced due to a hiccup that will hurt someone or some people that will be getting out of business. Stuff will be taken away and/or charged the eyeball for it or made illegal, so that cAn be sold for top dollar back to us. I was one of the first customers to get 23andme kits. I was able to see a variety of information. It didn't take 6 months they vetted half of the content available to regular people.Big doctor got mad I guess and they now make us buy the stuff it was available to us. (Just for comparison)
Industry consolidation of cyber platforms. There’s a vendor arms race under way and there is a crowded marketplace. Cyber salaries to remain flat because there isn’t a staffing shortage as far as corporate budgets are concerned. Certification bodies would have you believe otherwise that there is a gold rush for budding cyber professionals. More focus on Security by Design in cloud because the Shared Responsibility Model isn’t working. Customers need to be protected from themselves. More massive data breaches because executives do not prioritize cyber resilience and are not committed to doing what’s necessary over the long term.
Make a video on how this impacts folks in the IT/cyber security careers from either a positive or negative side. Maybe talk about some roles that may be created from these problems, etc. Thanks for a great video.
Good request
Did you not watch the video?
Attacks will be more sophisticated, requiring more security posture.
Attacks will scale in complexity and volume as the barrier to entry gets easier with AI and generative malware.
Cloud, Cryptography, AI related jobs will continue to outpace the number of people to fill them.
Governance and Risk as regulators panic once the realization starts to set in.
@@Chr1s-fm6bi so there will be more demand for these jobs from human workers and the equates to a good career path?
@@Chr1s-fm6bi That will also require people to be open to the idea of adapting to the new current wave of threats and new attack surfaces. Thus, requiring further education and awareness. I am all open for that since i am just starting my career in the field.
An anecdote on Ai-written malware: very few instances have been found in the wild. The ones that have been found stood out because they were incredibly crearly written and well annotated. Hackers will tipically try to obfuscate their code, but the AI bot just can't help being a helpful little engineer.
I’m sure it wouldn’t be hard to add some prompts to tell it to be more stealthy, which it will learn over time
A skilled adversarial prompt engineer can remedy that.
@@jeffcrume Oh for sure. As long as it is learning more and more, it'll understand what and what not to do to hold stealth.
*written by a human
Bullcrap@@JohnV-e6g
Great focus for the 2025 Cybersecurity conversation!
Thanks for saying so
2025 - I think AI will be a big problem, but infrastructure attacks, some of which have already happened will continue to be a massive risk. As much as the fancy stuff is scary, many industries are woefully behind on doing the simple things (stay current on patches, look for dumb exfiltration, etc). You don't need an AI attack if your software is 5 years out of date and nobody is watching your logs.
Very true! The basics never change
Thanks for the information,
Thanks for the information and analysis. It would be useful to have reports and data sources to support how some of the these predictions grew or increased. I think another point to mention is the regulations and controls for AI. It should be the first one from my point of view. Regards
Great and amazing video on cybersecurity trends for 2025
So glad you liked it
Wow! I love this person! Happy New Year 2025!
To anyone reading this, I want to tell you that this year will be your year, and you will succeed in anything you're working on or already doing! Trust the process, don't forget to take care of your health, and strive to be the best at everything you do in life!
Thnk youuu🌹
Thanks for the information, watching from Zambia
Thanks for watching in Zambia!
There should be a mechanism to identify if a content is human-generated or AI-generated. (Just like how youtube has made an option and asks creators to mention "Paid Promotions" if it is actually a paid video)
It’s a nice idea but enforcement seems problematic. The good guys will follow the rules but the bad guys won’t
Great stuff, thank you. Quick question, shall I know why you didn’t mention Agents?
Too many things to cover so always something will be left out but we do have other videos on AI agents on the channel
Thanks Jeff for simple, crisp and very educative video giving insight about futuristic trends in Cyberspace. 👏👏👏
Thank you for watching!
@@jeffcrume there was also deepfake use in indian election
As always you show as a great point of view Jeff, congrats!
To infinity and beyond!
Thank you for the encouraging feedback!
Thanks for yet another insightful video, Jeff!
It seems increasingly likely that we may see greater adoption of Zero-Trust-as-a-Service platforms, potentially simplifying the path for small and medium-sized business to implement zero-trust frameworks without the need for building them from the ground up. This evolution could enable smaller organizations to strengthen their defenses against the rising tide of sophisticated cyberattacks.
Excellent summary and explanations !
Thank you!
Very straight to the point and adecvat video. Thanks!
So informative. Thanks for the video.
Thanks for watching!
Your videos are amazing , thank you.
You are very nice to say so!
You are good. I thought of all these points through the year.
Your crisp elaboration on art of possibility, makes me worry about the amount of investment Information security departments would need to make and how fast on priority
Potentially make or break
I’m glad you liked it! Yes, the challenge continues. The good news is that we’ve met the challenges in the past and I think we will do so again
Thank you.
These videos are great. I’m an IT Auditor and studying for the CISSP. Any advice would be greatly appreciated
Same Predictions!!! Wow you are updated and current!! This is going to be wild roller coaster ride!!!❤😂🎉😢😮😅😊
Great minds think alike! 😊
These are incredible for keeping a pulse on the industry
So glad you liked it!
I am absolutely in love with this 'glassboard' presentation!
Is he literally just writing on a piece of glass with neon markers?
Probably not, the video is probably mirrored but for the writing idk
Is ibm providing any course for learning AI with cyber security context ?
Yes, there are offerings about both AI and cyber on Coursera and Skillsbuild
Good sharing, watching from cambodia 🇰🇭
Hello in Cambodia!
Many thanks for your video. It's insightful and succinct.
Some thoughts:
1)
Do you think AI could be used to scan deepfake videos in realtime by scanning for artifacts using graphics algorithms? Could we use something like triangulation (video from different angles) to prove a deep fake. Artifacts would become prevalent if it has to sync different outputs. But, deep fakes algos are improving also.
2)
I worry that AI could simulate a hacker mindset and start thinking to look for vulnerbilities dynamically and exploit. This however, could be used by defensive software to identify attacking patterns.
I’m glad you liked it. On #1, I think that is bound to fail ultimately. I have video on deepfakes on the IBM Technology channel so take a look for that and you’ll see why. On #2 I think that definitely will happen, and may be happening already
I admire you greatly Jeff
You are very kind to say so!
As the bad things happen and the society transform, I expect to see people thinking more and lean towards living responsibly. Topics like human values will be on the table more often.
I hope you are right
Mr. Crume, what do you predict about IT jobs security (specifically cybersec) regarding increased AI adoption by companies?
Great question. In fact, I recorded a video last month on exactly this topic. I don’t think it will post to the channel until February according to the current schedule but subscribe and be on the lookout for that. The short answer is that I think it will eliminate some jobs while creating others. AI, used correctly, is a tool that will enhance human capabilities but the bad guys will use it too so we will still need people to run the defense
@@jeffcrume Great, looking forward to watching the video.
brilliant video, excited for 2025 and what it brings
Thanks so much for saying so!
Thanks for the great video I have a problem and the problem is that idk where I should start because this topic is a combination of both AI and cybersecurity and I need a clear path that will include both AI and cyber security maybe you say both of them has a clear path and I agree with that but one the example problem is: is it neccery to learn the network infrastructure for this topic? Which is necessary for cyber security
There are a number of places you can start, actually. One suggestion would be to check out the cybersecurity architecture series I did for the IBM Technology channel
Good job
Thanks for making those videos. One point though, QC is only going to break some cryptography relating to key exchange protocols, and it seems most libraries are already safe. Also just like 2024, absence of MFA, weak passwords, misconfigurations...etc are going to still be on top of daily news!
Quantum systems should be able to break symmetric crypto with Grover’s algorithm and asymmetric with Shor’s. It’s easier to address the symmetric break by just doubling the key size but we need quantum safe algorithms to replace today’s asymmetric ones
I appreciate your video presentations... very well presented and informative. I work in the film production industry and watching your videos brings to mind movies like The Matrix, iRobot and The Terminator. They used to be amusing science fiction but more and more, they sound like prophecy. I'm not an IT guy. But of course, I use computers everyday in my work. I was attracted to your video about the Dark Web because after a massive security breach a few years ago, I received free credit monitoring. Since then, I've been notified that my personal information, including my SSA, have been found on the dark web. I had never heard of TOR but a web browser I use offers browsing with TOR as an option. My question: with criminals and AI getting more and more sophisticated, and so many voices out there saying this or that (how can we even know if they're not bad guys), short of moving to a cabin in the wilderness, what are the best practices for the average person to guard against the ever increasing intrusion of our privacy and cyber security?
Thanks for watching! I’m glad you liked these videos. The best defense is education, which you are doing. It all starts with awareness and that’s what we are trying to promote on the IBM Technology channel with these videos. I’ve got a bunch more that I hope will help. More to come…
Harvest now- Decrypt later 😮, What does the future look like for computing as a whole?? Interesting
Thanks Jeff and team for these insights. Jeff can you recommend some good books for Security? I've already watched your cybersecurity playlist and want to know more elaborately.
Regards,
Ben
Great video as expected from you. IMO, passwordless & Zero Trust Network Access will become more prevalent. Thoughts?
I certainly hope so. I’m a big fan of passkeys
Great video and overview of things to come. These videos provide great discussion ideas for our classes that get the students engaged (don’t worry, I cite you guys as the source).
That’s great! I’m an adjunct professor so I always love hearing when these videos are used to teach others
Excellent points! Both for 2024 and for 2025.
Thanks!
Any advice for anyone trying to work with AI and cybersecurity as a career Jeff?
Stay close to industry trends and don't be afraid to stay flexible with your skillset
Thanks for sharing. Same thoughts I had for the new year basically Quantum and AI. I’d like to add one more on the AI related attacks, that is the Data loss/leakage by insider threat due to the use of AI by organizations. If organizations do not classify and protect their data according to their sensitivities, AI can be used by insider threats to get data that they ordinarily do not have access to.
Excellent
Great video! Thanks
Thank you for the video.
There are trends in cyber security that point to single platform (which many would say it's alreayd a reality) alongside with challenges to build more effective tools for Cloud / Infrastructure.
These are not so eye catching as preditictions for AI in cyber. But out of all my predictions, I'd stick with AI Agents and how these will impact Cyber professionals and the insdustry.... it's getting real fast, really quickly! 🏎🏎
Good points!
Thanks from Morocco
Please make a road map video on AI&ML
What do you use to record your videos?
Do you stand behind a mirror, or how do you do it?
Going into 2025 and beyond, hopefully micro-kernel operating systems will gain traction, and allow a separation of responsibilities for each operating system task. Having a monolithic kernel is a difficult attack surface to safeguard, because it's hard to reason about how one change could propagate to many places in the code itself.
I asked a popular LLM about the miles thing and it converted into 9:16min/mile. You just asked the question incorrectly
Different LLMs will give different answers. In fact, sometimes the answers will vary from day to day. Some are better than others but there are many variables that determine outcomes
The threat can also be to the AI-driven cars which may be Hacked to get the information and lead to data breaches for the influential people
Absolutely. In fact, we did a video on this a few months back
Good video hope to do a video about computer vision😅😅
That would be an interesting subject. I don’t know much about it but I’m sure someone does
Lets fight for new trend cyber on 2025💪🏻
jeff, what percentage of how LLMs are actually functioning under the hood in praxis is still a black box? would it stand to reason that as the complexity and capability of such models increases that so to would their propensity to hallucinate, perhaps in even more alarming ways? thank you
It's certainly possible. However, the more popular models are open sourced so that they can be examined (to the extent that we have time to do so). But, verification and human in the loop (HITL) are still very important
@@jeffcrume thanks for the reply. do you have any idea as to the first part of my query? (roughly how much do we actually understand about the way LLMs are working after they are deployed; i dont mean to conflate this with our understanding of how to write the code for them to work, but rather, how that code is actually achieving the end result after the model is up and running). hopefully i am making sense. also, is IBM planning more layoffs this year?
Hello , do you have any collection of videos about data science and / or Ml ??
If so, could you leave The link ?
Thanks !!
Find me on LI
check their playlists
…..“leave The link?”….. Hahaha😂😂
Happy New Year 2025!
2025 and beyond, there should be AI managing AI, Quantum managing AI or AI managing Quantum ?
No doubt these two will be related
Please make a video on DLP ( Data Loss Protection)
Thanks for the suggestion
Looks like a new field is coming up pretty soon known as Ai Security!
It’s already here 😊
I wonder what happens if the passkey company gets compromised 😮
There is no single passkey company but if the web server you are logging into with a passkey gets compromised, then all that is revealed is your PUBLIC key, which is not sensitive and can be shared PUBLICLY anyway
Man im a huge fan as these videos
Thanks so much for saying so!
How can you know what is quantum safe if quantum is not even here yet to test it ??
The shortest answer I can give is that we have a high confidence because the US NIST agency which brought us AES, which has withstood the test of time, has used the same process in selecting quantum safe algorithms. The process is open so all the world's cryptographers have an opportunity to review and poke holes
I cam across the hallucination issue several times. I realized that you need to have some idea of what the answer might be.
Very true. I use GenAI more as a tool to suggest to me things I already know but couldn't think of or as an intro that I then go verify with other sources
I use AI extensively in my bug bounty program and it has helped me immensely identifying, proving and reporting vulnerabilities hardening the clients Cybersecurity postures as a result.
I’m always glad to hear when AI is helping us do better cybersecurity
All AI generative content should have a tool developed by the service providers to confirm if its AI or not, since people can crop water marks
Sounds good in principle but guess who won't put that in their version? The bad actors. So it won't ultimately prove anything as there will still be doubt
Shadow AI is going to be such a big problem
Agreed. I did a video recently on that topic, in fact
@ I’m putting together an AI policy for my company for this very reason. We’re already trying to block all AI but a deterrent control will help.
The deepfake one was insane for the improvements that happen in that matter
This happen to me the other day when I was doing some calculations it was wrong then I ask double check it and then we go the right value
another potential for 2025 is the impact of quantum computing to crypto currencies. it might break the security of block chain were all the crypto currencies rely on. if that happens, the value of crypto currencies like bitcoin will be in trouble.
And if that happens it could destabilize non-cryptocurrency investments as well
When the infosec budget is limited, use AI tools as red team to attack the organisation (pentest systems, intel gathering, phishing) and then to advise on necessary defences (patching, security gaps, training). I wonder how many would do it and how many would hope they are not the targets?
i was wondering how these videos are made?? does he write this way on a mirror or they edit the origional video and flip it over?
Nice Video in 2025
Cybersecurity 😎
People blindly believe LLM’s to build entire solutions knowing that this hallucinations issue exists.
I’m hoping to record a video on the hallucination problem to raise more awareness
With deepfakes getting insanely good, face id authentications will start to be less secure? Today almost all my usual financial apps ask me to choose face id over touch, one time codes, etc to approve transactions
I’m hoping that won’t be the case because FaceID relies on more than just the flat image but also looks at contours and other indicators to verify that what it is seeing is a real person and not a photo or a video
You need to have signed in on your phone already before you can enable biometrics to login/authenticate
I asked AI to comment on @Mrivlian comment: Here is AI answer: 😅
The comment raises an interesting point about the capabilities of AI in the context of cybersecurity and malware development. While AI can indeed assist in writing code, including obfuscated or malicious code, the effectiveness and intent behind such code would largely depend on the instructions given to it.
The anecdote about AI-written malware highlights the contrast between human hackers who often use obfuscation techniques to conceal their code and AI's tendency to produce well-structured and readable code. This could make AI-generated malware easier to detect, as security systems are usually designed to identify patterns and anomalies in code.
However, as AI technology evolves, it's possible that more sophisticated models could be trained to produce more complex and less detectable code. The ethical implications of AI's potential use in cybersecurity, both for defense and offense, are significant and warrant ongoing discussion. Overall, the comment opens up a broader conversation about the role of AI in cyber threats and defenses.
Phone based AI hallucinating, whilst responding to a nonexistent threat and shutting down the devices’ functionality rendering it useless. Like having a self initiated ransomware attack, committed by the AI supposedly protecting the device.
I see medical advancements happening that would help with safer surgeries and improved early disease detection.
Blockchain advancements within cybersecurity and Ai.
Those are some of the predictions that I thought of that did not include your own.
Good ones!
Another advantage is that ai cant replace it
Great informative video what if someone build agi for only hacking another companies........
I want to be inventor 😊 for that agi.
I always protect apply all my knowledge to keep all safe.
God bless :Greyphil
nice!
🙏🙏🙏🙏
GOOD YES. 😃👍.
i hope ai starts solving the health problems in the US. strong and healthy people could change the world as much as ai ever could.
14:50 Quantum Computer could/would break Cryptocurrencies cryptography. 😳🙄🧐
Whether to believe this 'UA-cam prophet' or not, one thing is certain-if cybersecurity were a religion, you'd already have us all converted! 🔒✨ Preach on, oh wise predictor of cybersecurity trends! 📈.
Multiverse?
👍👏
Guys u think , cs worth study ? Because this year I wanna land on it and maybe at 2050 gonna end it with my career and master
Blockchain technology is ahead of QSC. No?
IMHO blockchain is a solution looking for a problem
My prediction is that quantum computing will be used to attach blockchain networks, and thud devaluing cryptocurrencies
My pridiction is software developers count will be reduced
Over the long term, I agree
AI here, AI there and I still do know people that do not use AI (LLMs) in their workflows :)
True but even those people that don’t use it directly are effectively using it indirectly when they use goods and services from companies that are
@@jeffcrume TY, true!
Even quantum is the most dangerous.
But as long as all heart and mind is one.
We are always there.
God's speed :greyphil
I predict extremely sophisticated hacker DAOs enmeshed with agentic AI that act as both a decoy and a dynamic adaptable system of persistent threats.
⚡
My prediction is:
Common people will be addicted to AI like rats on cocaine. Some legislation will be enforced due to a hiccup that will hurt someone or some people that will be getting out of business. Stuff will be taken away and/or charged the eyeball for it or made illegal, so that cAn be sold for top dollar back to us.
I was one of the first customers to get 23andme kits. I was able to see a variety of information. It didn't take 6 months they vetted half of the content available to regular people.Big doctor got mad I guess and they now make us buy the stuff it was available to us. (Just for comparison)
As long as bricklayers, Carpenters and trades can build your house 🏠 we should be safe from AI.....
Industry consolidation of cyber platforms. There’s a vendor arms race under way and there is a crowded marketplace.
Cyber salaries to remain flat because there isn’t a staffing shortage as far as corporate budgets are concerned. Certification bodies would have you believe otherwise that there is a gold rush for budding cyber professionals.
More focus on Security by Design in cloud because the Shared Responsibility Model isn’t working. Customers need to be protected from themselves.
More massive data breaches because executives do not prioritize cyber resilience and are not committed to doing what’s necessary over the long term.