The biggest frustration for me regarding Claude for Enterprise is the pricing and credit consumption. Even for small changes, it consumes more credits than any other competitive platform. If there are more credits, the cost increases. I believe that is definitely a factor. Claude for Enterprise has done an exceptional job in delivering all the right things except for the pricing, because it comes with a good amount of cost. If it could ensure that small changes consume fewer credits, that would be helpful. My final thoughts on Claude for Enterprise are that with the growing use of automations, if Claude for Enterprise could come up with a no-code solution—currently it uses credits to code things—but if it could be a drag-and-drop type of solution, as other competitors such as Make.com or n8n offer, I believe that would be a huge advancement. Currently it is usable for even non-technical people, as they just have to use it with prompts, but if it were structured with a drag-and-drop method to create automation, it would be even more cost-effective and faster than coding it and finding the debugging parts.
The foundational model would have to be improved to be comparable to ChatGPT for everyday use cases. The system should be better at understanding when to reply directly with the foundational model and when to perform a web search.
The product could be improved by offering automatic integration with other solutions, such as the ability to read Excel or text files and automate processes this way. Currently, additional solutions are needed to integrate these types of tools. If I want to read a spreadsheet from Google and pose a human-like question, I need to integrate two or three AI solutions. Security is also a concern. AI solutions currently lack security in their processing. If I need to access my confidential information and the AI solution accesses and displays it in responses for everyone, it is not the ideal scenario. Developing solutions to ensure security is crucial, although it is not their main objective at the moment.
Claude for Enterprise streamlines workflows by enhancing code writing efficiency, accelerating task completion, and improving accuracy across diverse applications. It offers advanced AI capabilities, aiding professionals in boosting productivity while reducing time-to-market.Designed with a strong focus on workflow optimization, Claude for Enterprise aids in whiteboarding, debugging, and automation. It supports users with artifact features for data visualization and generates HTML outputs,...
The biggest frustration for me regarding Claude for Enterprise is the pricing and credit consumption. Even for small changes, it consumes more credits than any other competitive platform. If there are more credits, the cost increases. I believe that is definitely a factor. Claude for Enterprise has done an exceptional job in delivering all the right things except for the pricing, because it comes with a good amount of cost. If it could ensure that small changes consume fewer credits, that would be helpful. My final thoughts on Claude for Enterprise are that with the growing use of automations, if Claude for Enterprise could come up with a no-code solution—currently it uses credits to code things—but if it could be a drag-and-drop type of solution, as other competitors such as Make.com or n8n offer, I believe that would be a huge advancement. Currently it is usable for even non-technical people, as they just have to use it with prompts, but if it were structured with a drag-and-drop method to create automation, it would be even more cost-effective and faster than coding it and finding the debugging parts.
I face issues with Claude halting or hallucinating occasionally, as it happens with some projects but not all.
The foundational model would have to be improved to be comparable to ChatGPT for everyday use cases. The system should be better at understanding when to reply directly with the foundational model and when to perform a web search.
The product could be improved by offering automatic integration with other solutions, such as the ability to read Excel or text files and automate processes this way. Currently, additional solutions are needed to integrate these types of tools. If I want to read a spreadsheet from Google and pose a human-like question, I need to integrate two or three AI solutions. Security is also a concern. AI solutions currently lack security in their processing. If I need to access my confidential information and the AI solution accesses and displays it in responses for everyone, it is not the ideal scenario. Developing solutions to ensure security is crucial, although it is not their main objective at the moment.