In our orgnisation IBM watsonx.ai is primarily used to enhance threat detection automate security analytics and it helps to improve accuracy of incident triage within our soc Operations as security analyst I leverage this platform to analyze large volume of log and alert data to interpret them against threats and malwares or malicious behaviour from Qradar logs
Pros
AI driven LOG ANALYST and Investigation processing large volume of alerts and security events for incident categorisation and its identification that helps to detect attac at earlier stages of security incidents
Cons
while AI driven insights are accurate the reasoning behind alert prioritization or anomalu scoring is somtimes opaque analyst often need more transperancy into why specific event was flagged or how a confidance score was derived
Likelihood to Recommend
IBM watsonx.ai is highly effective in environment where multiple security data sources generate large volume of the data and alert wherein tool will be helpful in corlate login anomalies and lateral movement detection and data access patterns to identify early signs of incidents and its alert prioritizationd and false positive refuctions helps analyst to focus on genuine alerts
Well, in our business we use IBM watsonx.ai for different purposes. One of them being to help our customer support team respond to inquiries faster. It can quickly scan our knowledge base and suggest helpful resources so the manual efforts has been eliminated to a large extent. We are focused on automating support and improving accuracy.
Pros
It understands natural language questions, so I can ask things in plain English and still get answers.
Quickly scans through huge amounts of data
Integration is pretty well
Cons
Training it on our internal unique data took longer than expected
Could have a better intuitive interface
Customising responses for different departments can be tricky and sometimes require technical help
Likelihood to Recommend
We tried making an AI assistant/ chatbot out of it, worked and integrated pretty well with our process. We also use for summarising long docs, which saves a lot of time. However, the chatbot sometimes gave very generic answers, which I found to be a drawback. Also, another issue that I encountered was if the question was very complex, it started giving vague answers.
We use this to tune the llm, do prompting, deploy the models and then iterate for our multiple projects in medical domain like claims system, aba, chatbots etc. We are solving medical problems like medical record summarization, chatbot for QA from medical Record , ic code lookup and claims submission forms. It gives a good ecosystem of tools for miltiple usecases and all embedded within the same environment thats a great advantage and also using langchain is also great.
Pros
Integration with other systems
Deployment option within same exosyatem is great and can easily deploy any model .
Security layer for governing and hace insights to see the predictions and ai studio is also good
Cons
IBM watsonx.ai is expensive than other platforms.
Limited integraions though it has many but still some tools integrations not there for medical usecase
Its little difficult to learn as right now not many open reseouces
Community is not that strong to get any answer
Likelihood to Recommend
For genai apps its very good i can say where we don't have to worry about the whole ecosystem their whole ecosystem is flawless and very powerful analytical capabilities. It maintains the data Quality and data security. When cost is concerned and when there are large data involved. It becomes costly and tuning of model is not straightforward as there is no proper active community for which we can take help
We have used prompt lab, the granite models, and the slate models. Prompt lab was very easy to use, and the granite models provide many options as well provide benchmarking results which is useful in picking the right models
We use IBM watsonx.ai to build, fine tune and deploy AI models that directly impact how we plan routes and manage fleet efficiency. We replaced our previous multiple standalone scripts that didn't communicate as well with a centralized IBM watsonx.ai environment.
Pros
Autoprompt and tuning studio
A built in governance checking system
Its efficiency at training custom models
Cons
It's currently so hard to visualize trends beyond basic plots
Integration with non-IBM ML frameworks is quite patchy
Likelihood to Recommend
I'd say IBM watsonx.ai is 90 percent there, but advancing pretty fast. Right now, we use it to train custom models and it's really thriving. So anything custom models related will work so well. It's still struggling with managed scaling. If you can consult with expert firms, ours is Bluebay data, you'll make some really great strides.
From the Data Science area we use IBM watsonx.ai for PoCs with the business
DEsde El area de Data Science usamos IBM watsonx.ai para PoCs con el negocio
Pros
Recognizes invoices
Easy to prompt
Does not hallucinate
REconoce facturas
Facil de promptear
No alucina
Cons
It's a little high the price
Es UN Poco elevado El precio
Likelihood to Recommend
It's good for AI
Es bueno para IA
<i>This review was originally written in Spanish and has been translated into English using a third-party translation tool. While we strive for accuracy, some nuances or meanings may not be perfectly captured.</i>
In our organization, we use IBM watsonx.ai because we need a tool for help the all business , administration , optimization and create applications
Pros
create apps
optimization
automatitation
Cons
best implementation
more tutorials
best cost
Likelihood to Recommend
In my opinion, I would likely tell a colleague that IBM watsonx.ai is a tool to help persons but I think in a implementation in a business is very difficult, implementation for consults or tutorials in a system or erp
Just started a new project to implement the use of IA. So long seems to be really easy to use, hope everything keeps like this, We're looking forward to get more info and more use of cases to help my organization to get in the buzz that every body talks about,
Pros
Use of agents
Simplify tasks
Easy to learn
Cons
Show simple tutorials
Community support
More JavaScript alike
Likelihood to Recommend
I still don't have enough experience, but i've seen a lot of demos and i've made some real world scenarios and so far so long every thing looks fine, I came from Microsoft world and it's been kind of difficult to understand all the environment software and main frame focus
VU
Verified User
C-Level Executive in Information Technology (51-200 employees)