Log inSign up
Dr. Tejasvi addagada
697 posts
Dr. Tejasvi addagada profile banner
@TejasviAddagada

Dr. Tejasvi addagada

@TejasviAddagada
Tejasvi Addagada is a data, AI & digital leader assisting fortune 500 firms. I help build and optimize IT management and governance solutions.
Mumbai
tejasviaddagada.com
Joined April 2017
144
Following
51
Followers
RepliesRepliesRepostsRepostsMediaMedia

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
  • @TejasviAddagada
    Dr. Tejasvi addagada
    @TejasviAddagada
    Aug 2
    Every Organization Needs a Risk Atlas. Most Don’t Have One for AI and why do you need one" and more posts on the go. tejasviaddagada.com/article/every-… #AIGovernance #AI #Risk #Control
    Every Organization Needs a Risk Atlas.Most Don’t Have One for AI.
    From tejasviaddagada.com
  • @TejasviAddagada
    Dr. Tejasvi addagada
    @TejasviAddagada
    Jul 19
    The AI Guardrails in Financial Services: Why Independent Validation Cannot Be Assumed finextra.com/blogposting/32… via @Finextra #AI #Guardrails #security
    finextra.com
    The AI Guardrails in Financial Services: Why Independent Validation Cannot Be Assumed: By Tejasvi...
    If your institution has deployed a generative AI system in production, you have a guardrail layer. T...
  • @TejasviAddagada
    Dr. Tejasvi addagada
    @TejasviAddagada
    Jun 4
    Guardrails for GenAI in Indian banking: Why domain-native validation cannot be an afterthought etcio.com/s/lq6lxt6 via @ET_CIO. There is a need for adversarial benchmark datasets for Financial Services - often ignored while generalization hardly works #ResponsibleAI #finproof
    bfsi
    Guardrails for GenAI in Indian banking: Why domain-native validation cannot be an afterthought
    From cio.economictimes.indiatimes.com
  • @TejasviAddagada
    Dr. Tejasvi addagada
    @TejasviAddagada
    May 21
    General-purpose AI safety classifiers miss ~9 in 10 sophisticated prompt injection attacks in banking. We measured it. Semalith: ssrn.com/abstract=67287… #AIGovernance #BFSI
  • @TejasviAddagada
    Dr. Tejasvi addagada
    @TejasviAddagada
    May 14
    HarmBench, TruthfulQA, WildGuardTest are Excellent benchmarks — for general AI. Not one contains a KYC bypass prompt. Not one contains an investment advice elicitation in banking language. General benchmarks measure general models. Banking is not general
    Semalith v1.5: A Purpose-Built Safety Classifier for AI in Financial Services
    From tejasviaddagada.com