I tested the code scanning capabilities of 10 different, widely available, large language models (without cyber guardrails) using the same real-world codebase and asked: how good are the LLMs at finding security vulnerabilities?
I used the same methodology that I’ve been using to find 100s of vulnerabilities in open source software and libraries over the past month.
After model self-review, deduplication of the same findings across models and an independent model assessment, I found 350 distinct vulnerabilities.
The headline: on a single run, no model found more than ~35% of them, and false-positive rates ranged from ~2% to ~30%.
These were our results: 🧵
What a great line up at @ElevateTechFes this year! I’m so excited to be speaking at @ElevateTechFest this year. Join me on Monday at 11:30am in the #DataAnalytics track!
As said by our CSO @AlexRManea, "we don't just invest in software companies, we also build it ourselves."
We’re excited to share that we've partnered with Google's @TensorFlow Privacy to make our privacy-preserving software available to everyone!
Blog▶️ https://t.co/WtnhjTCxSk
A guide to tooling in Python ⬇️
Whether you're a Software Engineer, DevOps Engineer or Data Scientist this guide will help you set up your project
(or level up your skills) with the current Python best practices
https://t.co/ttMEzsC43h
#python
Hello twitterverse! If you are at @ACL2019_Italy, please come check out our posters on GANs using word embedding and an AutoNLP representation framework at #Repl4nlp workshop. We will be in Hall 4 from 3-430!
We are thrilled to be partnered with @BeamDental as they announce their $55 million investment round!
The company has continued to innovate and disrupt the dental insurance market and we are excited to work with the entire Beam Team on this new phase of growth.
Presenting our work on Unsupervised Discovery of an Educational marketplace at the AI for social good workshop at @iclr2019! Come say Hi to us at the poster session at 4:30 in room R05!
Our own Eddie Du speaking at @odsc East in Boston today. Eddie is an Applied Research Scientist on the Georgian Impact team and he's talking about Transfer Learning. You can read an intro to the topic by @azin_asg here https://t.co/vtX9M6bK7v
#ODSCEast#TransferLearningn
Excited to be attending @iclr2019 next week! @PariAIML and I will be presenting our work with @TopHat. Come say Hi if you are around! https://t.co/r1KiVHuFte
Thanks to @GeorgianPrtnrs for partnering with us to set up appropriate performance metrics and clearly communicate performance expectations to our users for our newest product #AuthorshipInvestigate 🔍 https://t.co/2KrZzxvXwI
We are thrilled to publish 'The 11 Principles of Trust', Georgian's latest white paper designed to help companies build a strategy to earn and safeguard customer trust. Download the guide here: https://t.co/YjeEIQ5cqQ
#trust
Trust is a balance between the value that you deliver and the
comfort that customers experience when interacting with your
company, brand and products.
Strike the right balance with the help of the CEO's Guide to Trust >> https://t.co/HczpBJgyG0 #guide#dataprivacy