“In the midst of chaos, there is also opportunity” – Sun Tzu, The Art of War – 5th Century BCE Today marks a significant milestone in the AI as DeepSeek, a Chinese AI startup, announced the release of its revolutionary R1 Open-Source large language model (LLM) rivalling OpenAI’s ChatGPT. This DeepSeek RI model has been designed to excel in complex reasoning tasks, rivaling the performance of OpenAI’s latest models while reportedly being developed at a fraction of the training & implementation cost. It is being widely reported that this R1 LLM was trained with Reinforcement Learning (RL) for a
Category: Algorithms
AI, Algorithms, Artificial Intelligence, Big Tech, LLM, Machine Learning, Software Development, Tech Leadership, Tech Trends, US Federal Government Oversight, US National Security, US Politics, US Regulators
Beware of Human-injected left-leaning bias emanating from AI Large Language Models (LLM) Outputs – RLHF technique could be the misused
In the realm of Machine Learning, Reinforcement Learning with Human Feedback (RLHF) stands out as an innovative technique where human trainers play a crucial role in guiding the learning process of models. Unlike traditional reinforcement learning, which relies solely on pre-defined rewards, RLHF incorporates human judgment to shape the training environment. This method can have significant implications, especially when it comes to ensuring that models consistently favor certain outcomes over others. In this blog, we’ll delve into how trainers can influence models using RLHF, highlighting both the potential benefits and pitfalls. Human trainers can introduce biases, whether consciously or
Personal Privacy – A Mirage in today’s Tech World? Renewed call to the incoming administration to protect us!
Are we giving up our privacy for convenience without thinking about the consequences? It seems that we all desire data privacy, but our actions often seem to indicate otherwise! We have always been calling for users to be careful of what they post online. My kids are familiar with my saying to them all the time “think before posting / texting / tweeting anything online… because once you do, it will stay there forever!”. It is our view that people do not fully appreciate how pervasive our online digital footprint is and the amount of information (related every aspect
Strategic Countermeasures to combat Software Vulnerabilities effectively in AI/ML enabled applications
Looking back, Application Security has evolved significantly in the last couple of decades. In the early 2000s, SQL injection and Cross Site Scripting (XSS) attacks were a nightmare for cybersecurity teams as attackers easily bypassed network firewalls through attacks at the application layer. Since traditional network firewalls at that time were not application-aware, these attacks proved a blind spot allowing attackers to compromise web applications easily. Hence, the computer industry developed countermeasures which included and not limited to web application firewalls (WAF), source code security reviews, and DevSecOps, who automate these checks within CI/CD pipelines to and allow security
Algorithms, Big Tech, Cybersecurity, Tech Leadership, TikTok, US Federal Government Oversight, US National Security, US Trade Policy
TikTok Takedown Order: National Security recommendations to prevent actionable surveillance of American Citizens by a Hostile Foreign Power (China)
(Uploaded 8/25/20) TikTok is a very popular app with over 1 Billion downloads, which translates to almost 1/7th of World Population. This app is very popular among youngsters here in the USA, just as it has taken the world by storm. TikTok like all other “free” social media platforms, treats its users as their product, their personal data as its currency, and their Business Model is predicated on trading one’s personal information to use these “free” services. The app’s privacy policy says TikTok “may share your information with a parent, subsidiary, or other affiliate of our corporate group” while