aiisjustanif
Joined 8 years ago
Joined 8 years ago
You can try model all you like, but eventually you have to go argue with finance because you work at Snowflake and your new SaaS product needs a new SKU and is slightly different than this other offering because it has enhanced security on larger storage.
Don’t forget there is a new manual process for troubleshooting this because it overlaps with another feature potential that we found in testing. You need to go make sure your Customer success team is aware of this and has at least some internal documentation.
Ah crap your sales director says they had 2 other competing features that take priority over your feature to train up Sales Engineers to cross sell. So you have meet with existing customers to try and garner interest and also work with comms to make another product release post with higher visibility.
Oops do forget to grab your next ticket for next sprint that just started from Jira, since you are a dev too.
This problem existed before AI and great when you caught it.
There’s always been a percentage of luck getting offers from big tech companies, but it’s way more luck than than merit it has ever been.
> What bad PR?
It literally changed Apple's marketing and privacy features for years to come [1] and at the next request of the FBI forced them to choose privacy and security as something the would like to be known for and incorporate into their products long term. [2] Snowden made a note how effective Apple's encryption had become in 2016. [3]
> Why would you even remotely believe that they are not under a similar situation with the current admin? It's just ridiculously naïve.
Multiple aspects of the larger program was done without the knowledge of the tech companies, PRISM was no where near the level of pervasiveness as Upstream. [3]. They inserted a man in the middle on the fiber lines leaving data centers to copy traffic, functionally making backdoors unnecessary.
> 3rd party contractors had access to your siri conversation
I'm glad that one was leaked, it's something that a lot of us thought was happening with many voice assistants, and it shows just why sensitive quality control should not be outsourced.
> Your position is just not credible in 2026 . . . ignores two decades of history. . . Reading such argument on HN tells me Snowden did throw away his life for nothing, damn.
Same goes to you, you ignored all improvements related to this discourse over that time. I'm highly skeptical of many corporations, including Apple, but my goals are to understand the whole picture and direction. Overall Apple has been up and down in terms of security and privacy commitments, but significantly better compared to companies such as Google, Huawei, and Meta.
[1]: https://www.apple.com/apples-commitment-to-customer-privacy/ [2]: https://en.wikipedia.org/wiki/Apple%E2%80%93FBI_encryption_d... [3]: https://www.youtube.com/watch?v=cJ6PpX6xg-E&t=1808s [4]: https://www.eff.org/pages/upstream-prism
AWS enclaves [2] have been around for a few years now with decent research on the topic and they are highly secure from personal experience and pentesting [3]. The idea at least is similar for exclaves, instead of a separate VM or a completely detached coprocessor chip, Exclaves create isolated resources on the main system-on-chip.
[1]: https://randomaugustine.medium.com/on-apple-exclaves-d683a2c... [2]: https://aws.amazon.com/ec2/nitro/nitro-enclaves/ [3]: https://arxiv.org/pdf/2206.04123
Any CISO or head of security loves the ideal of completely using AI to handle incidents, tune detections, implement mitigations, track vulnerabilities, pen testing, etc. This feels like it driving security teams to have less critical thinking.