-
Here are some MacOS tests from AV-TEST.
As you can see, most of the ones tested are pretty good. I recommend further research on review sites if you want to tell them apart. Something to check is if they also scan for Windows malware, as that’s helpful to have.
Malwarebytes is excellent for on demand scans on Windows, but doesn’t offer full system scans on Mac.
Something else to bear in mind, if you have an M1/M2 Mac, is to check if the AV software in question runs natively on these chips or not. I remember looking into this about a year ago and being surprised that some still run through Rosetta, which is obviously going to be a performance hit. Some do run natively on M1/M2, though.
1 user thanked author for this post.
-
-
To quote my post on the subject from October:
Just because antivirus software can’t access the kernel on MacOS that doesn’t make them useless. They can still detect malware and block it in the user space before it does harm, not to mention that good antivirus software will also detect Windows malware before it can spread to your other machines. MacOS malware is at an all time high and should be taken seriously.
People wouldn’t create two million pieces of malware in a year for MacOS if it is naturally immune to malware! And the Windows machines that you presumably interact with with your Mac certainly aren’t immune to malware either, so it’s good to protect them in advance.
$2000 on a Mac and then skimp on $30/year for antivirus software that takes up 1% of its CPU and memory? That doesn’t make sense to me. As I said on another thread, about Linux antivirus software, which also applies to Macs:
Is it likely that you download malware that targets desktop Linux? If you’re a smart user then probably not, but it’s not about being smart, it’s about the fact that you’re mitigating a low risk high cost scenario with a low opportunity cost activity (running an antivirus), so from a risk management perspective I still think that it makes sense to install one.
-
-
It’s impossible to prove that this is the case, but given that Bing AI creates its answers based on things previously written, and since humans have written so much sci-fi about sentient evil AIs, when Bing AI is prompted to explore itself I wonder if it might be calling on these sources.
3 users thanked author for this post.
-
-
My reference to detection rates was based on the work of AV Comparatives, where Microsoft falls behind most of the big names when it comes to Real World Detection.
It is curious that the security.org article suggests the opposite, though, I agree. I wonder if this is an example of sampling bias – that is, maybe people who pay for AV software are more likely to encounter viruses than those who don’t? I assume that they don’t weight for the number of viruses encountered. Interesting puzzle.
1 user thanked author for this post.
-
As Microfix and Alex said, I think that most people are just happy to use something for free rather than pay for it, even if tests show that the detection rates are slightly worse.
Not to mention how some of the various paid-for options try to trick users into paying more and more. I recently helped out someone who had bought Norton for £20 and they asked her for £85 for the renewal. And then when she paid, she found out she had been charged twice!
Paid AV software isn’t a monopoly any more, and if they treat their customers like that then they’re going to lose them.
1 user thanked author for this post.
-
I understand that my post could have been taken as criticism of your post rather than the article, and so I apologise once again for upsetting you and for not being clearer. I hope that you can accept my apology and understand that I was just trying to help.
As you say, a tempest in a teapot.
3 users thanked author for this post.
-
-
I think that geekdom’s point was that AI isn’t the agent here, and semantically the thread title makes it sound like it is.
For example, if someone used a computer to commit fraud, you wouldn’t say that ‘Computers are after your bank account’.
The trick that the phase ‘guns don’t kill people, people kill people’ uses (not that I’m even remotely saying that you’re trying to trick people!) is related to the fact the the word kill has multiple meanings: it uses kill to mean ‘consciously kill’, and uses the absurdity of the statement to imply that that ‘guns kill people’ must be false, even though ‘guns kill people’ uses the word kill to mean ‘increase the killing of’, not ‘consciously kill’.
Similarly, the phrase ‘AI is after your bank account’ implies that AI is consciously after your bank account. Something such as ‘AI usage increases fraud’ would instead clear up the ambiguity.
I hope that this makes sense.
1 user thanked author for this post.
-
Google indexes around 50 billion web pages.
Facebook has 17 billion posts and 10 billion messages every day.
There are 350 billion emails sent every day.
Wikipedia has over a billion edits.
YouTube has around a billion videos.
Just some examples.Do you really think that it’s practical to have a lawyer decide that each one of these web pages/posts/emails/etc. complies with every US law before it is published?
Do you really want to wait for a lawyer to review your email or transferred files before you can access them or send them?
And if you extend it to them being liable for data stored, as you have said, that would mean that the services would be liable as soon as something was uploaded to them, meaning that all of the above would be impossible to exist, as would any form of data uploading, since all it would take is one bad actor to upload an illegal message or file for the service to be at fault before the lawyer had even had a chance to review it.
1 user thanked author for this post.
-
-
-
Author
