Amazon Relents, Lets its Programmers Use OpenAI's Codex and Anthropic's Claude (futurism.com) 12
An anonymous reader shared this report from Futurism:
In November, Amazon leaders sent an internal memo to employees, pushing them to use its in-house code generating tool, Kiro, over third-party alternatives from competitors. "While we continue to support existing tools in use today, we do not plan to support additional third party, AI development tools," the memo read, as quoted by Reuters at the time. "As part of our builder community, you all play a critical role shaping these products and we use your feedback to aggressively improve them."
It was an unusual development, considering the tens of billions of dollars the e-commerce giant has invested in its competitors in the space, including Anthropic and OpenAI... Half a year later, Amazon is singing a dramatically different tune. As Business Insider reports, Amazon is officially throwing in the towel, succumbing to growing calls among employees for access to OpenAI's Codex and Anthropic's Claude... Given the unfortunate optics of opening the floodgates for Codex and Claude Code, an Amazon spokesperson told the publication in a statement that teams are still "primarily using" Kiro, claiming that 83 percent of engineers at the company are leaning on it.
It was an unusual development, considering the tens of billions of dollars the e-commerce giant has invested in its competitors in the space, including Anthropic and OpenAI... Half a year later, Amazon is singing a dramatically different tune. As Business Insider reports, Amazon is officially throwing in the towel, succumbing to growing calls among employees for access to OpenAI's Codex and Anthropic's Claude... Given the unfortunate optics of opening the floodgates for Codex and Claude Code, an Amazon spokesperson told the publication in a statement that teams are still "primarily using" Kiro, claiming that 83 percent of engineers at the company are leaning on it.
What will happen (Score:2, Insightful)
Re: (Score:2)
The idea seems to be that nobody will train more AI, ensuring the entrenched players with already-trained AIs maintain exclusive control over those domains.
Re: (Score:3)
I'm not sure this result is inevitable.
In November I decided to rebuild a personal web project in Vue.js. I didn't know Vue.js, at all, so I got Copilot to help me set it up and start rebuilding the app. In the process, I personally *learned* Vue.js. I can now write components with no help from AI, because I observed what AI was doing.
AI *can* be a knowledge *building* tool, if we have a mind to learn. Those who forget how to program because they use AI, probably shouldn't be programming.
Really? I wonder (Score:2)
I have tried AI(today's automation products) for coding and what I end up with is very familiar examples across different tutorials I have seen that don't compile.
Don't get me wrong, I have programs that will take a field list and kick out a c++ database class I can start with in my app and that works great and saves time.
But that is automation, not AI and I have been writing programs to crank out code segments since the 90's so this type of automation is not
Re: (Score:2)
I think it's both. I've personally gotten a lot of use out of claude recently just for quickly getting started with somebody else's code (we weren't even allowed to use it a month ago even if we wanted to, which I didn't until I was specifically asked to use it.) E.g. ask it a question like "where is X done?" or "where should start for working on Y?". I don't ask it to make any direct changes. Basically the kind of stuff you do with a knowledge transfer, only you don't have access to the original developer(
Re: (Score:2)
I think what you are pointing at is that AI has a closed world assumption built into it. The reason is that it cannot know your context, but without that context, one doesn't even know of what a solution would consist. By closed world assumption, I mean it seems to operate as though it had all the relevant information and can confidently proceed on that. Put another way, it doesn't know "that it doesn't know" (not "what it doesn't know", that would be a contradiction in terms).
If the above is correct, we ca
Sounds familiar (Score:4, Informative)
That is a classic corporate "vanity project" trap. Itâ(TM)s a common mistake: a company mistakes its captive audience (employees) for a test market, forgetting that internal tools need to solve a problem, not create a new chore.
I worked at a company that tried to force an internal social media site on us. The catch? They weren't a social media company. It never took off. It was a textbook example of why you should stick to your core competencies.
Re: (Score:2)
The problem at AWS is that they largely don't have 'core competencies' anymore, and haven't realized it yet.
They used to be a company which embraced new ways of doing things and doing small, agile things quickly. That hasn't been the case for half a decade now - in part due to cultural changes pushed from the top, but largely hasn't been the case for a while.
You'd think a cloud company with a fully distributed global infrastructure would have been one of the forefront proponents of remote work, and they did
Re: Sounds familiar (Score:3)
Kiro isn't bad, but it's not the best. Allowing them to use the competition will be good. It'd be like Microsoft not allowing its employees to use Macs, at some level you need to be exposed to the competition to actually be competitive, because you can't float by on your customers being naive forever.