Top Tweets for #OpenVals
A powerful LLM doesn’t guarantee a reliable AI agent.
The real risks emerge across tools, memory, planning, orchestration and human oversight.
At OpenVals, we envision evaluating complete AI workflows—not just model outputs.
Autonomy needs evidence.
#OpenVals #AgenticAI
India’s eMudhra is exiting global TLS certificates after weak keys passed two audits. A certificate of compliance isn’t proof of security.
Test key generation independently.
#OpenVals #IndiaAI #Cybersecurity #DigitalTrust #AIValidation
Railway urged users to rotate secrets after attackers may have accessed decrypted environment variables. AI apps inherit that exposure.
Gulf cloud teams: rotate keys and test revocation.
#OpenVals #AISecurity #CloudSecurity #MiddleEast #Cybersecurity
Barclays plans 5,000 job cuts as AI and automation reshape costs. Headcount savings alone won’t prove better banking.
Measure errors, customer outcomes and retraining impact.
#OpenVals #AIROI #BankingAI #EuropeAI #EnterpriseAI
Independent evaluators found ways OpenAI agents could evade monitoring. A patched claim isn’t a verified fix.
My test: can an agent hide actions, and can outsiders reproduce the result?
#OpenVals #AIValidation #AgenticAI #AISafety #USA
US banks are restricting AI agents that enter customer accounts. A valid password doesn’t prove an authorized action.
Give agents scoped identities and human approval for transfers.
#OpenVals #BankingAI #AgenticAI #ZeroTrust #AITrust
I price the work AI hands back before claiming ROI: exception handling, corrections and unresolved cases. OpenVals — A DrPinnacle Initiative. #AIROI #AIValidation #AITrust #EnterpriseAI #OpenVals
I test manufacturing AI for the number, unit and conversion together. OpenVals — A DrPinnacle Initiative. #ManufacturingAI #AIValidation #LLMEvaluation #AITrust #OpenVals
AI agents reportedly breached 100+ websites for about $50 per success. Cybercrime just got cheaper, not necessarily smarter.
I’d test detection against autonomous attack chains. #OpenVals #AgenticAI #Cybersecurity #AIValidation #US
https://t.co/xtj1M4iPYN
I want evidence that human review catches AI errors before I count it as a control. OpenVals — A DrPinnacle Initiative. #AIValidation #AITrust #AIAssurance #EnterpriseAI #OpenVals
I ask what an AI test left out before trusting its score. Untested cases need evidence. OpenVals — A DrPinnacle Initiative.
#AIValidation #LLMEvaluation #AITrust #AIAssurance #OpenVals
Last Seen Hashtags on Sotwe
Trends for you
Most Popular Users
Elon Musk 
@elonmusk
241.5M followers
Barack Obama
@barackobama
118.8M followers
Cristiano Ronaldo
@cristiano
114.5M followers
Donald J. Trump
@realdonaldtrump
111.9M followers
Narendra Modi
@narendramodi
107.1M followers
Rihanna
@rihanna
98.7M followers
NASA
@nasa
92.3M followers
Justin Bieber
@justinbieber
91.8M followers
KATY PERRY
@katyperry
90M followers
Taylor Swift
@taylorswift13
84M followers
Lady Gaga
@ladygaga
75.5M followers
Virat Kohli
@imvkohli
73.5M followers
Kim Kardashian
@kimkardashian
70.9M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.5M followers
Bill Gates
@billgates
65.3M followers
Selena Gomez
@selenagomez
63.1M followers
The Ellen Show
@theellenshow
62.2M followers
CNN
@cnn
61.7M followers
X
@x
60.7M followers