Bright Ideas Agency is an independent consulting company, owned and operated by Nick DeCourcy, helping small businesses supercharge their productivity through digital transformation. This is a channel for business owners, individual information workers, and IT professionals to help find out about the tools and techniques we use, and our learning along the way. Our focus is predominantly Microsoft 365 and the Power Platform.
Disclaimer: We can only offer specific technical and business advice to businesses or individuals we have an existing business relationship with. The content of this channel is general information, and while all effort is made to provide correct details, no individual or business should rely upon it without seeking professional advice related to their specific situation. We are not accountants or lawyers, so do not offer financial or legal advice.
Its's an interesting viewpoint that indemnification makes it ok to use tools built on mass copyright violation- but I see a downside here- what happens if the next IP to be casualy appropriated is your own? If we all agree that it's ok to ignore IP rights as long as we can 'get away with it' are we not creating a world where no one's IP is safe from similar treatment? To use tools built on stolen IP in the hope of creating exploitable IP would seem to be the very definition of building your house on shifting sands.
The idea that customers must stop using a company's products because a claim of infringement has been made against them doesn't stand up to scrutiny. New York Times and others who have brought claims against OpenAI have one perspective on their use of content, and OpenAI (and clearly Microsoft too) have a different one. As to my view, I've shared it in another video ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-kaHFzo3RlKw.html The fact is that there is constant litigation going on about who stole IP from someone else. Going back to 2022 HBR was reporting in a general IP infringement issue in big tech (hbr.org/2022/08/big-tech-has-a-patent-violation-problem), and you'll recall that just recently Apple had to pull infringing watches from the market when it was found guilty of infringing another's patent. The expectation that businesses don't deal with the products of companies that are being sued for infringement is not sound. Ultimately, every business decision we take has risk associated with it. Imagine if you were a software creator who decided to build a tool leveraging Apple's infringing technology. You'd be out of luck. Fundamentally your comment paints indemnification in the negative, but this is not reasonable. As a business owner, I buy insurance to create indemnity, I adopt contracts that offer indemnity; indemnification is the means through which we can de-risk reasonable actions we take in business that ordinarily would create greater risk. If a company with the scale and bank account balance of Microsoft is willing to provide indemnification for your use of their product then this both helps de-risk it, but also should signal that they believe their legal position is sound.
issues I've had are: negation ( prompts which ask specifically not to include something), consistency between refinements - the models don't keep things the same between successive refinements.
Yes, getting gen AI to exclude something is far harder than it makes sense for it to be. The lack of consistency is really a feature though, because a new response is generated each time without reference to the last, you can end up with substantial differences in how similar prompts are responded to.
Many people who never or seldom use AI tools or dismiss them quickly are disillusioned and negative, only consuming opinions that feed into their confirmation bias, Many of those will not change their minds about the current generative AI tools, no matter the amount of change management. But I think many of them will use AI if it is integrated into existing apps without a separate AI label, like “Generative Remove” in Lightroom or machine learning in Amazon search. It may also help if Microsoft removes the “Copilot” name and makes AI more seamless in M365 to reach more users.
I agree with removing Copilot. A name used everywhere to mean different things is the same as no name at all, it’s just confusing. IMO Copilot could be a feature added elsewhere, but it shouldn’t be different in every app.
Hi Nick, appreciate your point of view. Certainly some tools are a benefit, such as summaries and composition. Some are really at an alpha, not even beta level. I have subscribed to both Google’s Gemini as well as copilot 365. The area of most frustration relates to limitations of data handling. I did a test this week using copilot and Gemini. I have three years of quarterly electricity accounts in both outlook and gmail So I prompt the respective bots to tabulate the last three years invoice amounts in a table so I can load into excel or Google sheets. Copilot fails miserably but Gemini will tabulate 18 mths worth 6 invoices but no matter how I prompt it will do no more than 6 lines or citations. But it does provide a decent table. Therein lies the rub, it shows promise but at this time it’s limited. The best results come from careful curation of the files you have the bot ingest and carefully prompt to get good answers My take is that it will take time and patience, maybe in 18 mths 3 years the issues will be sorted. So it’s possible to see real benefits but it’s limited today unless you are enterprise and have staff addressing the limitations.
This is an interesting use case but one that is arguably not a great fit for LLM-based generative AI. I would probably approach this type of need in SharePoint then using a form processor or Autofill column to focus in the detail of each file with AI rather than trying to do an aggregation initially. I have found all the main AI chat tools to be fairly unreliable with type of workload. Undoubtedly some of these things will get better but I think that ultimately we will gain better understanding of what type of model is best suited for a task, and likely the scope of a tool like Copilot will expand by offering different types of models or logic approaches.
@@brightideasagency it’s a simple test of the state of AI service. I did some more digging and even asked Gemini 1.5 how to improve the promoting then ran the suggestions only to get a final answer to use Google gmail search and look at each result 😳 I think there is an imposed limit of only six citations- both MS and Google have a way to go on anything not straight forward and likely to be useful IMO
I agree. To deliver on the promise of these tools there’s some way to go. However, I’m finding that by piecing together more traditional document automation alongside tools like Copilot there are ways to get a lot closer without a lot of extra effort. I have come to frame this very much like we might think about getting the best out of a team member. Do we focus someone in on trying to improve their weaknesses or do we find ways to play to their strengths? When we play to Copilot’s strengths and put it in a team with different tools that have other strengths the end result is much more complete.
I wouldn't say they underdelivered, the features are simply not done yet. It's super smart and a ton of fun to talk to and use, but the extreme stuff is coming, just a little later.
I appreciate your perspective, but imagine a similar scenario applied to a different type of product. "Here are the keys to your new car, the windshield you saw in brochure isn't ready just yet, but we'll let you know when you can come back to get it fitted". In that case, you would say the car manufacturer underdelivered, no matter how pure they original intent to deliver the whole car. If the product you told people you'd give them isn't ready don't ship it, better yet, don't announce it until you're in a good place to get it out the door on time.
There is definitely a trend toward the more chatty unless you alter the default. However, in some uses I appreciate this, particularly when chatting with ChatGPT on my phone.
Yes. Ultimately, I think we can expect the experience will be aligned with the demo when the new voice mode arrives. However it’s definitely later than expected at this point and the app as it stands right now is less useable than we probably expected based on those demos.
@@brightideasagency I whole heartedly agree - but I assume it was to stick it to Google a day before the Google I/O and they rushed it a bit. I have come to the realization that much of this promised land is still on the horizon and not an actualality. I do agree that they all should probably tone down the over promising and under delivering at this point. But it's a maturity game at this point and within 10-12 months we will see many of these things arrive, if not even more crazy features once the next gen of models rolls out. What I am more worried about is the knee jerk reaction from US Tech Giants to not roll out in Europe becaue of the DMA and AI Act Regulations. This is a much much bigger issue for Europe as a whole and our competitiveness in the coming age. One which I don't see enough focus on - many of the senior MDs/Country Managers I speak to all seem to think that the EU regulative space is a competitive advantage - and I really don't see it this way at the moment.
Aren't all these issues connected though? The stupid overhyping of these products and rushing them to market before they're ready is what has gotten regulators spooked. You have Sam Altman telling us this is the most impactful technology ever developed, and then his company gets into a spat with Hollywood actress over whether or not they coopted her voice. If you're dealing with a serious tool, you need to behave seriously, and focus on communicating accurately. The failure of the tech companies in showing they are up to the challenge of being mature and reliable stewards of this technology is what is creating a questionable regulatory landscape.
well you have a point. But in general I think we are over-regulating the digital space in Europe. The DMA regulation is not directly related to AI and is the primary reason they are pulling the plug on EU. When it comes to the AI act, I don't think our law makers are even remotely equipped to fathom much less regulate the area of AI. And then they try to preemptively regulate a space, which haven't seen any major incidents yet based on fearmongering. Google not launching their models in EU for 8-12 months and now Apple not launching Apple Intelligence will be a major hit for our competitiveness in the long run. It is a very sad trend we are seeing and something which will have major implications for us in general.
4 months, everyone in the OneNote subreddit was bragging about how Loop was going to add automation to OneNote. It really looks as if all it does is things that OneNote already does by itself, but in a way that forces users to make use of their OneDrive. It's the same kind of crap that Microsoft fold in OneNote when OneDrive first came out. And, everyone complained enough that Microsoft went back to allowing the free version to access notebooks stored on the desktop. My guess is that they will eventually do the same thing with this loops thing. You will be able to put a Loop component into a page in a notebook that is stored locally. The component will need to be stored on OneDrive, but it can be displayed in, updated in, and synchronized with the component that is cached locally.
The reality is that many modern experiences in M365 rely upon cloud based storage. It’s understandable that Loop is limited to the web as at its foundation it’s a platform build for collaboration.
@@brightideasagency Funny. For a business that has, you know, actual servers, all the IT dept. has to do is create a shared folder, stick a OneNote notebook in that folder, tell everyone where it is... And BINGO! Collaboration. OneNote literally does an excellent job of synchronizing the cached notebooks of everything using that shared notebook. That was the selling point of OneNote from very early on. If Microsoft's focus truly was "collaboration" they could let you store a cloud, based notebook on any of the cloud services, and let OneNote take care of everything itself. But, some executive figured out that that was a mistake, because it left people less reliant on OneDrive, and less likely to continue paying ongoing fees. But, I get it. If you say anything contrary to the Microsoft taking points, they will cut you off from early access. So, I know you have to pretend that everything Microsoft does is "for the customer." That any limitation is just due to the way the technology works. Completely ignoring that Microsoft made sure that the technology would work that way.
11:30 If you don't trust microsoft ... then the logical recourse is to not run windows, and while you'r at it, don't use microsoft 365 either. Exactly, I avoid microsoft as much as possible. But my children are not able to do this because their school demands it. And I imagine a lot of people need to use it at the workplace.
In a business or education setting the organization needs to assess how the platforms they use meet their objectives. In the education sphere at least schools have had tech for a long time that makes it inadvisable to be using those shared machines for sensitive personal information.
Hi there. I have a question on the Region and time zone settings. I changed this to my current time zone and checked the box, "always show time slots in business time zone" We do business nationally, so should I leave off that checked item? I would love for the user to view times available in there time zone wherever they are booking.
Yes, you generally want the guests to be able to view the time slots in their local time zone. Bookings will make sure they get on your calendar at the right time.
Great demo. I remember seeing a demo of syntex at a Microsoft workshop 5 years ago. But it has never found it's way to any orgnizations i've worken with. I hope to see it finding more attention, because this is really cool
What a wank I think you just demonstrated that Syntax is a completely useless product for a non enterprise business Why would you bother with such outdated stuff SharePoint et al has always been useless and MS is trying to now make it relevant My take is that you need to reimagine document storage and retrieval - surely with AI we can have a blob of documents and you should be able to pose a question at the blob and get the best response - isn’t that the basic promise of MS copilot?
As with so many of things, I think the answer is "It depends". Like any system reliant upon data, Copilot suffers from a certain amount of "garbage-in, garbage-out", and if you have really disorganized, repeating, or contradictory data, Copilot is going to do a much worse job for you. There are also issues of recency and scale (in relation to the data extracted for the individual request) for Copilot that mean for certain tasks it just isn't currently well suited to extract the right information reliably. In my view, Copilot and Syntex can live harmoniously with one another and are complementary. If the data/files you are working with require structure and repeatability in how you work with them, then Syntex works reliably for this. If what you are doing is new, ad-hoc, or creating content then Copilot is a great tool. And increasingly we'll probably see the two sets of tools come closer together as is the case with Autofill columns. If your point is "can't we just point AI at a bunch of documents", my push back is "of course, but which documents?", and for many businesses this is the fundamental challenge of retooling with AI, and where having some kind of document automation capability comes in.
@@brightideasagency thanks Nick, I understand your view but my experience with Sharepoint in geeneral is not good, probably due to my own impatience to understand its features. Its 2024 and MS should be promoting a fresh approach not surfacing old technology and spouting old saws
I did find this VERY interesting as I am "currently" a unemployed IT professional, working mainly "help desk" that last 12 years, and I have accepted employment as work "outside" of IT. I think having a good "overall" knowledge of different areas of IT is great to know. I am currently studying for COMPTIA Sec+/CySA+, (only because I have vouchers), but I have a recently passed CompTIA Cloud Essentials+ and interested in "Cloud", but looking at Desktop Admin, to get employment in the IT area, as I am Windows 7 Desktop Admin Certified. But must get current, so these fundamental Certs would work great..
Thanks for watching. Yes, they are a good starter into a new area. Good luck with the certs you are studying and getting the type of role you're looking for.
If you brought someone from 50 years ago to today and showed them how we deal with sharing information, our money management, purchases etc, they would think it was crazy. Giving others our data is a double edged sword, it definitely has downsides but it also powers many of the technologies we use today from turn-by-turn directions to online banking. Will this end up being no different? Time will tell.
It appears that you are not familiar with Apple's development tools. You continuously pushed the idea that you have to use all of Apple's apps in order for Apple Intelligence to know what you are doing. Let me re-introduce you to Spotlight. As an iPhone user yourself, I know that you are familiar with it. But you seem to be so steeped in Microsoft's philosophies that you assumed Apple must do the same. Apple developer's all have the ability to send their data to the Spotlight search engine, if they choose to. That means that any third party app can add to Apple's AI personal context. Add to that Apple's App Intents and developers can give Apple's Siri as much or as little control to manipulate and use their applications capabilities as they choose. Contrast that with Microsoft Copilot where they do seem to be keeping most if not all AI possibilities with their own 365 apps. As to security, I agree with you to trust, but verify Apple's claims. So far, they have not betrayed that. Google's search engine deal has never to a secret to my knowledge. I agree that it is odd that a supposedly privacy first company has a widely known privacy violator as their default search engine. But that is a topic for another day. At least you can pick any other search provider you choose. In fact, with AI I expect that Apple will finally have to unveil its long rumored in-house internet search engine-both for privacy and for additional data for its own AI efforts.
Thanks for this input. That’s an interesting set of information on how search is already architected across Apple’s devices. However, I will still be interested to see what 3rd party app support this enjoys on day one as if they were able to integrate across their entire ecosystem why not show that? Frankly, if Apple demonstrated an ability to fully index my personal data based solely on what apps I chose to install AND do that without security compromise, then my video would probably just be titled “Copilot Killer” as I would regard that as a solution far better for most users than what Microsoft is currently trying to do. I am certainly open to the possibilities of what Apple has laid out and will definitely be trying this out on my iPhone and Mac when it becomes available.
Excellent video, shame you haven't explored this further. Understandable why, but I feel more people are looking for the solution as this is a useful option.
🎯 Key points for quick navigation: 00:00 *🚀 Initial Steps for Co-Pilot Adoption* - Understand the purchase requirements, licensing, and options for Microsoft 365. - Define the target user base and identify the why behind adopting co-pilot. 03:40 *📊 Data Preparation and Management* - Importance of data quality and alignment for effective co-pilot responses. - Considerations for data storage, sharing, and access rights within Microsoft 365. 06:11 *🛠️ Leadership and User Engagement* - Reflecting on generative AI tools and ethics for leadership understanding. - Importance of fostering a community, communication, and engagement around co-pilot adoption. 09:10 *📚 Training Strategies for Co-Pilot Adoption* - Tailoring training programs to cover practical use cases and AI principles. - Leveraging internal power users and external resources for varied training needs. 11:14 *💡 Individualized Support for Co-Pilot Users* - Recognizing the need for personalized help for executives or specialized users. - Strategies for providing one-on-one assistance to enhance the user experience. 12:39 *🌟 Long-Term Co-Pilot Adoption Strategies* - Emphasize the importance of continuous adoption efforts and evolving with the technology. - Reviewing progress, making course corrections, and exploring new areas of integration for co-pilot. Made with HARPA AI
Within ChatGPT there is documented file-type support help.openai.com/en/articles/8555545-file-uploads-faq and it does not seem to expose site contents such as pages. In Copilot for Microsoft 365 these file types are listed as being indexed within the Copilot Semantic Index, but they are not exposed as files you can add via ContextIQ when you type a / in any of the apps or Copilot with Graph Grounded Chat.
Microsoft has responded to early concerns about this by implementing just in time decryption. This represents Microsoft’s insider software releases working. Experts have got their hands on the software, provided feedback, and Microsoft has responded to fix prior to general release.
When you create the flow, do it by creating a rule in Loop. It will populate that information for you. I’m not sure how you would go about getting this information manually, and, ultimately, unless you create a rule in Loop there is no ruleid.
It's a problem for a simple reason that you have to ask yourself, "How well do you trust Microsoft or any other company to not push the boundaries of what they're able to get away with." For one, Microsoft doesn't have the best track record as is, knowing this, the biggest problem is having to take their word for things, they say "This data will be stored locally on your computer and not be sent to the cloud," I can tell you right now, I don't believe that for a second, a tool like this is obviously meant to be invasive, cause most of the pros are so minimal, but the down-sides are just off the charts and this entire feature is literally just spyware that is just being disguised with the whole AI craze nonsense and there is no way something like this is just going to be stored locally at all or if it is, it won't be that way for long. Add to the fact that this is already half-baked with their lack of content moderation where all your credentials and everything are going to be on display, I even watched a video where this is also going to include stuff that's on your browsers (including edge, even though they mentioned that browser was going to be exempt from the recall feature, which was clearly a lie it seems), he put in some credentials of his own (of course fake to try it out) went to recall and all the information was right there for everyone to see, oh but guess what? I also heard that if the recall feature has anything related to the company themselves, any other company, or some special interest groups, that information will be disregarded immediately and maybe to some extent could be censored somehow, so "If" this is going to be the case, then that really shows you that even Microsoft knows that this is a really invasive feature so much so that they have to exclude themselves from it so there is no sort of liabilities on their end, but oh! The customers get to deal with this "great feature" and we're just exposed to the wild, fascinating. With the whole hacker thing as well, sure you could definitely argue that they do have other ways of obtaining your information that doesn't have to go through recall, but the problem here is that, now they just have a 1000x easier way to get CRAZY amounts of information about you all at once, and it's all packaged nice and neatly where they can just search through whatever they want, how is this "not" a problem? And this is even worse if the person has gotten remote access so they can just grab what they need and split and you're done for. Also, let's keep in mind that the government and others will likely have a field day with this feature now and considering how bad censorship is these days, now anytime you type anything on your browser you have to be super wary of, made a joke that might be seen as offensive and recall captured it? Could get you in trouble, are you a journalist and there are going to be people out there that want to take you down? Recall captures it. Are you an anonymous figure online that makes RU-vid content like for example "Vtubers" where this is their strongest asset, well now recall can have all that information right there for everyone to see, and again they "say it's going to be stored locally' but we all know its not or at least for long, and if some kind of data-leak happens when it does get stored on their servers well, your fucked and they won't have any liability over it either. Then, let's add on top of all of this where, this feature is going to be turned on "automatically" by default, so a person that doesn't even know about this feature could have all of their information just shown to the world without them even knowing, certainly isn't creepy at all, and then to make matters worse after every update, they turn this back on, so if you happen to forget to turn it off after each update and you just go about your regular day searching through stuff then you're kinda done for, and sure you can "delete" these snapshots, but it does it literally every 3 seconds (someone said 5 once, but I'm pretty sure it's 3) so it's kinda useless. And sure you can toggle this feature off going to settings>privacy and security>recall and snapshots, but how long is that going to last? And how do we know that even when you toggle this off, it's not still working in the background? This is one of Microsoft's biggest feature yet, do you really believe them to let you just be able to turn it off that easy? No way. This entire "feature" is just a surveillance camera for your computer to monitor your habits online 24/7, and while it's true that this won't be rolling out for every system out there, cause of the requirements are going to be quite extensive to run this and there is a certain processor you have to have being something called the "SnapDragon" so most windows 11 PC's won't have to worry about this "for now", (although they do already have it where you can download some feature or whatever where it allows you to run recall with minimal specs requirements and seeing as they've already done this, you know they have this planned to be rolled out at some point for EVERY windows 11 PC and future PC's down the line). Seriously though, the fact that these guys haven't gotten some kind of class-action lawsuit at this point is kind of baffling to me.
Clearly this feature needed more work, but I think some of the commentary is overblown. To my understanding no one demoing the limitations of Recall was actually doing so on a system approved for it to run on. That’s like misconfiguring your router on purpose and complaining when someone gets into your network. I’ll be interested to play with this in an insider build.
@@brightideasagency At this point the Copilot+ PC's have already rolled out and there have been people that have tested it on those systems and it's more or less what I've already described. Although, word has come out that Microsoft is dialing back the feature (not completely getting rid of it sadly) and going to polish it up, cause they got so much back-lash not only from the community, but cyber-security experts, but considering they didn't even bother to do their due diligence a feature like this was made to be thoroughly secure from the beginning, I don' t have high hopes there will be any massive improvements and while of course you are going to have a very niche amount of people that will have a use for this feature, this is something that should have never seen the light of day really. But, have fun testing out the feature, if there's any interesting bits that you end up finding out about it, I'd definitely like to know.
I really think that recall is cool. I would love an AI to do that personally. But not Microsoft AI that's built into my operating system. Plus Microsoft has proven they will 100% force this to turn on in the future with updates. If I used an AI like this, I would want it open source because I'd not touch any closed source AI that was so invasive without honestly being able to see the code. This is such a massive security and privacy issue. I was already really struggling to move to Windows 11 for multiple reasons. It was a hard pill for me to swallow, but when they announced recall, I bit the bullet and moved to Linux. I've used Linux for years, but always for server work. But as of a few days ago, I'm official on Linux! And holy cow Linux for desktop use has come so far since last I checked in! Games work pretty decent too! I'm setting up a Windows VM with Glass Panel for games that need Windows, but other than Adobe products and games, Linux is honestly superior in every way from what I can tell. Like legit everything else. And the gaming aspect has come a long way. I honestly wonder if in the next 5 years, Linux will have little to no issues with games. But I'm honestly kind of shocked Microsoft thought this was a good idea. Even if they back peddled from here, this is a Unity situation but worse. I am a Csharp developer. I am a Csharp die hard fan boy. But, after Unity pulled the stunt they did, I would be weary ever touching Unity with a 10 foot pole. Still solid arguments for it, but you now know the evil and desires of Unity and the direction they wish to go. What Microsoft is doing is 10X worse than that. This is such a horrible invasion of privacy and security. And they seem to have zero desire to back peddle. But even if they were like, "Woops, are bad, we pulled this from the shelf". Is anyone going to trust they won't slip this back into a future patch? Because if you think they won't. You're high lol. This spooked me off Windows completely. Damn do I miss the days of Windows XP. Windows 7 was also solid in my eyes. Windows 10 was a worse version of Windows 7 imo with a constantly crashing file system, but it was fine enough I guess. Though to Windows 10's credit, it did do a vastly superior driver support job. But near the end of Windows XP. But for many many moons Microsoft has been spyware and placing backdoors in their OS. But you could fight it, plus it wasn't always the biggest of deals the way they were doing it. But now, this is just next level. This is sooooo scary.
If individual users enjoy Linux, great. There’s lots of places it’s useful. But it’s only been a few years away from takeover for two decades or more at this point and it still has practically no users in comparison to Windows. I might even try Linux again myself, but I’m pretty sure most of my small business customers have zero interest.
As an IT admin, I stay up to date with what Microsoft is pushing, that doesn’t mean I like it. Just started dual booting in Nobaru. This IS a security/privacy issue. Anyone who has local access to your machine can access this data. What about a library who doesn’t know how to properly deploy a system with group policy? The grandmas and grandpas who fall for scams where a scammer remotes into their system? Absolute insanity. Sure you can say the same thing about a password manager, but it’s not a screenshot of what you’re doing every day, every 3 seconds. This has personally put me off of windows in a massive way, and I believe many, many power users feel the same way.
The only vaguely awkward thing for me about Recall is that if your PC gets a virus / hacked... Or the governments gets its hands on your PC... They're potentially going to be able to get a lot more information. The core concern behind privacy has always been that you don't want your data in the hands of people who are actively likely to harm you... Corporations have little incentive to harm you - too much trouble for their profitability if they end up harming a user. Anyway, this is ultimately part of the question of how Personal AI agents should work... I'd personally consider your own agents as a part of your digital identity and deserving of legal protections against unwanted search and seizure, etc. And access to such systems should ideally be more strongly locked down, so that even if your PC is compromised, that too much isn't leaked outside... Which shouldn't be that hard to do. Although it does mean that Copilot should automatically not remember certain kinds of information. It might also be worth it to completely separate business context work from personal context... Like, you might only use Recall while your laptop is in business mode (depending on your preferences)
It is an option. First, if you don't want it, don't buy a Copilot+ PC. Second, if you have a Copilot+ PC and you don't want it, don't activate it during set-up. Then, if you've turned it on and change your mind, turn it off. 100% optional.
This is such a terrible take, I really expected better from you. Journalists are pointing out real practical concerns that have nothing to do with the integrity of the operating system, and everything to do with the operating environment that computers currently work in. They are not always locked. Passwords are not always secret. They're not always using individual user accounts. Not all developers have all their API keys hidden all the time. If you're not thinking about the practical implications of such a large change in the security model of an existing product ecosystem, and instead only pointing out the technical aspects, then what even is the point of an agency? You might as well just redirect to Microsoft's marketing. Seriously, I would reconsider this perspective, look at it from a real-world, human situation, and perhaps consider taking this video down. It's not all journalists and all industry that are stupid, it is in fact Microsoft that have done something so obviously bone-headedly inhuman here, that the technical aspects are not the point.
Thanks for watching, I appreciate your considered response to my perspective. I think you make fair points on the range of concerns a feature such as this may interact with in a negative way, but I still think it reasonable to highlight a counterbalance to much of the reporting in this issue that both neglects to highlight the real potential benefits of this innovation and the fact that this does not sit alone and isolated as a solitary attack surface. Ultimately, I expect Microsoft will take feedback on this preview product and improve it, but we should remember that initially at least this will be on a vanishingly small number of PCs, owned by those who have selected a device specifically to get access to AI features such as these.
I think if you don’t trust Microsoft with Recall then don’t use it. If you don’t trust Microsoft at all, then don’t use Windows or Microsoft 365. There are definitely options. But the idea that any significant number of users of dropping these platforms and shifting to Linux en masse is just not reality. I don’t think there’s a reasonable argument to suggest Linux is the safest option for the average user, for power users and enthusiasts it can be great.
I absolutely agree with you, Fundamentals exams offer diversification of your technology knowledge and shows your interest in learning for growth to prospective employers if your have litrle or no IT experience, however those with some experience in IT, it's probably best skip Fundamentals exams and go straight to Associate and Expert level certs
0:12 yeah Unix lalready fixed that over 30 years ago, every heard of the commands: grep, find, locate, recoll, history,etc.. Heck if those at microsoft are so computer illiterate at least they could learn to press ctrl-H to search their browser history. And what about Rover the dog in windows XP from over 20 years ago!! Rover found any and all files by name, content, date of of last modification, etc.etc.. so you can't remember what document you were using yesterday? pop up Rover on windows XP and bam! you have it instantly and with less than 1GB of RAM. Remember only one word inside the document but don't remember the name of the document, Bam! Rover found it instantly! Why did microsoft screw up their search and just about everything on windows since windows XP?
I appreciate your input on this, but I think you are focusing in on a very limited aspect of the feature set that has been displayed in Recall. AFAIK there is no similar search capability in any major operating system that indexing content based on the semantic meaning of what is displayed in a document or on screen versus just indexing the text in the file. You may well be able to find a file based on remembering a word, but you most definitely have not been able to find a file based on a "graph with purple text" in Unix/Linux for 30 years. It's a fair criticism to say that Windows search has degraded, I don't think it's fair to not recognize the innovative experience in the Recall solution.
I think most of a fear mongering has been over the top. Recall at best is taking Random screenshots so most data will be missing, password fields on web sites don’t display passwords so they can’t be captured. If a hacker/malware has so much access to your pc to view the Recall data, then they will also have full access to all your files, web browser history and passwords etc. I would be much more worried about the full access at that stage. Most of the online fear I have seen has been based on things that haven’t happened yet e.g. MS sharing your data, but that’s a different conversation if that ever happens. For everyone using facebook and google/ chrome, you have no privacy, all your data has been sold. Recall in its current form can’t share anything about you unlike other popular services. If you are worried your parents/wife are going to view what porns site you have visited, don’t turn on Recall and clear your browser history :)
Thanks for watching. Totally agree. I understand there being concern, but a lot of people are behaving as if this feature arrived in a vacuum where this is the only potential risk that has ever been created.
Thanks for putting up a balanced and well thought out video on this. I’ve had this debate on a number of sites about Recall and it hasn’t been pretty. Guess most commenters just want to bash Microsoft. As a software developer with 30+ years on the Microsoft platform, if I can get enough access to get to Recall database on your computer, Recall would be the least of their issues. Since at that stage it’s only a short matter of time before I have full access to all your banks, pensions etc via other better and more reliable attack vector’s. I think a lot of the ‘outrage’ is also a lack of understanding on the basics of how computers work and what your real security risks are. Also can’t help but wonder if Apple had released this product what the reaction would have been :) Anyway keep up the great content.
@@MnemonicCarrier even then the password isn’t guaranteed to be captured. It someone has access to Recall, then they also have full access to all your passwords via your browser or password manager and everything else on your pc.
Scam. They change the A.I. info. Seen it myself. Asked a question one day and asked the same question the next. It was changed in favor of the mega companies.
Sure. AI does not "know" things in the way humans do. I recently released video on this topic, check it out if you haven't already: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-2ktYkWjCodM.html
The thing is, as me many people work on offices, and in my case, I have a particular workload where when its need I do my part of the work and some other time I watch video on RU-vid or read things, and other things not related of my workload, but its because I don't have demand, I bet in the feature this Recall will be available for admins to check their employes and this sure will be a nightmare, I am not doing anything wrong, but if an admin check my Recall and see I am watching videos about games, how I should explain this?
We have seen before and will again that with digital tools to track or oversee work come new challenges in workplaces. For workers the question comes down to whether what you are doing that you want privacy for is defensible as reasonable and is being protected by you for reasons other than obfuscating how you are spending your time. For employers the question comes down to what balance of trust and oversight leads to the best results, not just in terms of productivity, but also in employee retention and morale. If you're working on a business owned device, then the best assumption is always that your employer can see 100% of what you're doing, we don't need Recall for that.
If I hadn't switched from Windows to Mac since 2004 and owned one of those new Windows AI PCs, the first and only thing Microsoft's "Recall" feature would record on my PC would be installing Linux and saying goodbye to Windows.
Every time I see someone crying about MS "spying" they're an iPhone or Android user who have no problems with Apple and Google recording every word they say, face scanning every photo they take, and scanning every document on their phone to upload to iCloud and GoogleDrive. Hypocritical virtue signalers pretending to care about privacy, when in reality they're just fanboys hating on their chosen enemy corporation while taking it up the *** from their favored corporation. 1984 already happened 10 years ago on iOS and android. Apple and Android users have already been arrested for photos they took getting misidentified as CP by their AIs and automatically reported to the police. Out of every "person" I've talked to that's flipping out about this ALL of them have been iOS and Android users, zero exceptions. Clown world.
Nothing is totally secured... so it is a privacy and security nightmare even if it is also an inovation. But for me, like the browser, it should be an application, not a part of the operating system. An operating system should be only a software layer allowing an application to run on a computer: a file system, a memory manager, a network manager, a GUI but nothing the user directly interact with. Of course you can have default application but I think that it should be explain during the installation and you should be given the choice to install them or not. And for the "Innovation" part: if we cannot remember anything without AI assitant, if we cannot make any innovation without AI assistant we're gonna end in idiocracy... limited to the innovation the IA was trained on... If you want your brain to work you have to make it work, not rely on AI assistant.
I think these are fair points, but the reality is that this separation of roles is not how operating systems work. One direction those wanting to switch from Windows hold is Apple but that’s an even more integrated walled garden than Windows. As the completeness of systems out of the box has increased the complexity for users has decreased, little time saving additions like being able to open a Zip file in the OS rather than install a 3rd party app actually makes usage safer. Choice can be great and secure for those with knowledge, but for the average user it introduces complexity and the potential for making decisions that are less than optimal in terms of the security of their system.
@@brightideasagency "not how operating systems work" : well it was and and it still is for some. When you install Debian you have a lot of choices. I love the "politicaly correct" style of your answer. My translation is "let's assume that the average user is not so bright and don't want do understand how a computer works but just want it to work out of the box. Instead of trying to educate, let's impose the microsoft software and suggest that any other choice may be unsafe". Windows and MacOS are both software made to lock customers for profit with argument like: we known better than you what you want. Personaly I use a 14 years old HP workstation. when I bought it, it was running Windows NT. I updated it to Windows 7 but it was the end for me with Windows, Linux gave me the freedom I wanted . It can run Windows 10 but not Windows 11... not problem with the last Linux. Of course the storage was changed a few time (Came with SAS drive, changed to SATA SSD and now NVMe SSD) but I didn't need to buy a whole new PC to install the last version of Linux, need one if I want to install the last version of windows...
I think it's unfair to jump from installing Debian to "the average user is not so bright". The days of computers being solely for people who are interested in them are long gone, and if you're spending your time installing Debian either you work in IT or you have a great interest in IT as a hobbyist. I also enjoy technology for the sake of technology, but I respect the fact that a lot of people don't, and the desire the average person has for their PC to simply work without having to learn a great deal or tinker continuously is no negative reflection on their intellect. It's great that there are products to allow such a diversity of people, interests, and learning styles to gain benefit from technology in their life and business, and the same platform will not be the best choice for everyone.
@@brightideasagency The last Debian is very easy to install, using all the default options it's not more complicated than windows. "the same platform will not be the best choice for everyone" : I can't agree more! But a customer using GAFAM products usually don't understand how they work: they are easy to use and he don't look any further. A video like "Data Brokers: Last Week Tonight with John Oliver (HBO)" should be mandatory to watch before using any device connected to internet. And this video was before using AI. People need to be aware that if they use these products they take the risk of loosing any privacy. They should be because in many case it's written in the "terms of use" but who reads them? And no, don't expect big companies to respect their customers, you can find many example in the video of "Louis Rossmann".
The Linux golden age has been coming since I was in college. The reality is that a large number of people already have Linux derived devices around them (Android, ChromeOS) but it seems unlikely a single Windows feature is going to move the needle on enterprise use of Windows. In many ways the recent years have shown the fragility of the open-source supply chain that Linux relies upon to a far greater degree than Windows, so I'm not sure either platform is without its problems. If some users want to move to Linux in response to Windows 11's development direction, then that makes sense, but I doubt I'm going to see it in too many offices soon.
my issue is that people keep booking then rescheduling minutes before the interview is supposed to start. How can I make it so they need to connect with me to be able to reschedule? Any help is appreciated!
Thanks for watching. No, there is no explicit capability to do this. The point of using a tool like Bookings is to allow users to self-service manage their appointments. However, I would suggest adding terms to the effect of "Reschedules with less than X notice will only be honored in exceptional circumstances. If you need to reschedule at late notice please email Y to provide an explanation, otherwise your appointment will be cancelled and you will no longer be considered for Z". You could also attach this to some sort of Power Automate flow to run the late notice change through an approval workflow so you can automate removing it from your calendar. Assuming these are job interviews, I would probably just take what you're outlining as an intolerable red flag about that candidate's reliability without explanation.
