Igor Khalatian, Ph.D.

Just finished watching Pavel Durov's interview with Tucker Carlson. It was a captivating dialogue between two titans whose impact on both journalism and the tech industry is profound: Pavel, who co-founded "VKontakte" while he was a student and later pioneered the messenger app Telegram; and Tucker, known for his insightful reporting across various esteemed publications and networks. This interview, crafted for a Western audience, sheds light on Telegram's strategic direction, particularly its innovative revenue-sharing model enticing content creators. Yet, I found myself yearning for insights into Telegram's financial strategy—Pavel hinted at complexities in this area, a testament to the challenges of scaling a global platform. According to publicly available sources, by 2026, Durov needs to repay bondholders over $2 billion or conduct an IPO, a decision he appears reluctant to make. Exploring the mechanisms Pavel intends to leverage to address this dilemma would have been enlightening. Moreover, Pavel's revelation about an unexpected encounter with FBI agents during a breakfast meeting in the US piqued my interest. Given the backdrop of his cryptocurrency venture and the subsequent SEC investigation, one can't help but wonder about the ramifications for his business. Overall, it was a riveting interview, offering valuable insights into the intersection of technology, media, and entrepreneurship. Highly recommended for anyone passionate about the evolving landscape of digital products and platforms. #Durov #telegram #Paveldurov #GemSpace

1

A question I often get from customers is what tools should we be using, and how should we assess whether a tool will help us? Kostas Hatalis, Ph.D. covered this in depth with Fenwick & West's Vejay Lalla in this article. It's a great read on how to get into the DETAILS of a tools capabilities and serving your business. Check it out at the link!

2

Thrilled to enable 💪 Bedrock - llama 3.2 cross region inferencing - https://lnkd.in/dad-phUS s/o Steffen Ridderbusch (+4 more updates 👇) 💸 Azure AI rerank endpoint cost tracking s/o Emerson Gomes 🛠️ Logging - response redaction on `turn_off_message_logging` s/o Ilari Scheinin 🔨 Cohere - support cost override for cohere rerank models s/o Jerome Roussin ⚡️ Proxy - fix returned headers 

8

Embracing Ethical AI in Medicine: Key Insights from GIHF-AI's Latest Policy Briefing The latest policy briefing from the German Israeli Health Forum for Artificial Intelligence (GIHF-AI) highlights the ethical principles and trustworthiness essential for the successful integration of AI in healthcare. Here are the main points: 🔹 Trust and Transparency: Crucial for public acceptance and effective use of AI. Trust is built through transparent guidelines and continuous quality control. 🔹 Human Autonomy: AI should enhance, not replace, human decision-making. Ensuring that AI supports human autonomy is key to ethical AI use. 🔹 Continuous Quality Control: Regular updates and certifications are vital. This ensures AI systems are reliable and perform as expected in medical settings. 🔹 Data Management: Proper handling and legislative support for data management are necessary to maintain trust and protect patient privacy. 🔹 International Collaboration: Balancing innovation and regulation globally is essential. Countries like Israel, with advanced AI applications, contribute valuable insights to the global discourse. 🔹 Human Responsibility: Medical professionals must critically assess AI-generated results. Ensuring that doctors retain ultimate responsibility for patient care is fundamental. At Innoccu, we are committed to following the regulations and directives of the relevant authorities to create a solution that is also ethically oriented. For a deeper dive into how ethical AI can revolutionize healthcare, check out the full article here. 💡 #AI #Healthcare #Ethics #Innovation #GlobalCollaboration #TrustInAI https://lnkd.in/dk6a-GNK

1

Another interesting article commenting on two different court cases related to AI use (one is the already discussed rejection of AI-enhanced video evidence in Washington state). "Until courts can appropriately understand how results from the use of artificial intelligence (AI) are generated and the sources relied upon by the AI program, there will be a great reluctance to accept such results as supporting the point sought to be made in court. Recent decisions addressing AI-enhanced evidence and AI-generated results demonstrate that courts want to know what is in the “black box” and, without understanding the “black box,” courts are unwilling to accept AI-created results." https://lnkd.in/edXGh7QB

27

Please join me on Tuesday, April 16 at 12pm ET for a conversation about deploying artificial intelligence technologies in your IP practice without stepping into the numerous pitfalls that exist. I will be joined by Jason Harrier (head of patents for Plaid), Asian Shaikh (partner at McDermott) and Austin Walters (founder of IP Copilot and inventor on 500+ patents). #ipwatchdog #ai #artificialintelligence #webinar

11

We will keep being in the middle of the discussion on #GPAI #AI to help get the #AICode right. More #EuropeanAIRoundtable coming up! There’s too much posturing and fearmongering, too much blatant protectionism and anti-AI bias in the discussions. Our role at #CCIAeurope is to work not just with our membership but far beyond to come up with workable solutions that can lead to a successful #Code despite the many obstacles in the coming short few months. CCIA Europe #digitalEU #tech #BetterRegulation #innovation

3

1 minute to illustrate the incredible power asymmetry in access to AI hardware, and how Kalavai helps unlock computing for those who need it most. #gpu #ai #crowdsourceai

23

🚀 Unlock the Power of Custom LLM Evaluators! 🔍 Are off-the-shelf LLM benchmarks falling short for your specific use case? It's time to revolutionize your approach! Key takeaways from our latest video: 1️⃣ Custom evaluators are the secret sauce for domain-specific LLM applications 2️⃣ Our feedback approach requires 20x less data than traditional methods 3️⃣ High-quality evals lead to high-quality data, boosting LLM accuracy 4️⃣ Say goodbye to expensive human labeling - hello automated improvement! Don't miss out on this game-changing insight for AI teams. Watch the full video to learn how custom evaluators can transform your LLM strategy. #AIInnovation #MachineLearning #LLMOptimization #DataEfficiency What's your take on custom LLM evaluation? Share your thoughts below! 👇

17

Hope folks can join me at MIT Technology Review's signature AI conference -- EmTech Digital on May 22nd, where I'll be sharing experiences and thoughts about the various AI litigations currently filed and what might be ahead on the 12:20 session named "AI Goes to Court" (please don't @ me with "but AI doesn't have legal personhood so your title doesn't make sense"...I didn't name the talk). And if you're there, feel free to pull me aside and tell me why you think I'm wrong in my approach. Always interested in feedback and sharpening my arguments. https://lnkd.in/eRUndMvw

15

“The Justice Department has charged two employees of RT, a Russian state-controlled media outlet, in a $10 million scheme to create and distribute content to U.S. audiences with hidden Russian government messaging,” The indictment PDF (in comments) is full of info on the operations and how it worked! I'm still digging through it in more detail, but some initial takeaways: 1) Background on RT and targeted platforms: This includes platforms like TikTok, Instagram, YouTube, X (formerly Twitter), Facebook, and Rumble. 2) People involved: Kostiantyn Kalashnikov (username: kostya_k on Discord) and Elena Afanasyeva, both employees of RT, were the main figures on the RT side. They used shell companies and sock puppets such as "Eduard Grigoriann," "Helena Shudra" (e.g., afalena1997 on Discord), and "Victoria Pesti," with the latter sharing an IP address with Helena. 3) Leveraging influencers: The campaign used organic YouTube influencers with large followings, unknowingly convincing them to produce content aligned with RT's mission. 4) Financial operations: Nearly $10 million were allegedly laundered through a complex network of shell companies across multiple countries. "Following the money" is its own interesting deep dive in this indictment. 5) Coordination: The indictment mentioned the use of Discord for communication between Kostiantyn and Elena. 6) Impact: The operation managed to produce and disseminate approximately 2,000 videos, garnering over 16 million views on YouTube. This content was designed to be highly engaging, maximizing viewer reach and influence, and significantly extending RT's covert messaging. #osint #doppleganger

17

1 Comment

#SLM based #AppliedAI toolsets like SCOTi from smartR AI are mindful of Privacy People and Planet as well as purse-strings. https://lnkd.in/eAuMzAws

11

#AI agent benchmarks are misleading, study warns - https://buff.ly/4ePB09p #AIagent #LLMs #genAI

1

🔭 In a thought-provoking turn of events, #Anthropic underscores the potential perils associated with AI systems, calling for a well-structured regulatory framework to nimbly avert AI disasters. 🏗️ The organization vehemently voices that targeted regulation is not just an option but a necessity, to harness AI’s advantages while concurrently deescalating its inherent threats. • Logic and math capabilities • Superior coding skills • Ability to reason 🧩 As AI systems evolve in these powerhouse capabilities, the potential for misuse remains latent, especially in critically sensitive sectors like cybersecurity. Thoughts to ponder: 🤔 ✔️ Will a well-thought regulatory blueprint assure that AI's potential does not morph into an unintended Pandora's Box? ✔️ In mis preemptive measures such as regulation, can we ensure that innovation is not stifled? In a digital era, where we are heavily reliant on machine learning and AI systems for various aspects, it's inevitable that this resilient technology will permeate across all echelons of our society, from healthcare to automotive sectors and beyond. Interestingly, is AI's expanding capability a one-edged sword? Or is it the double-edged samurai sword that holds power in one hand and risk at the other end? This conversation resonates that: 🌠 Fear of the unknown is a fear, but so is the fear of the known when we don't control it effectively. Balanced foresight and smart regulations can indeed create a safe & robust AI frontier. Are we ready to face and embrace the conundrum? #ThoughtLeader #AIRegulation #DigitalFrontier.

2

On-device capabilities are the future of many AI use cases. Privacy, cost, offline functionality, and reduced latency will all play crucial roles. Check out our latest blog on how to run Small Language Models (SLMs) and build RAG pipelines on devices. #SLMs #RAG #AI #GenAI

28

AI isn't just for tech giants anymore. It's revolutionizing work across industries, from finance to energy. Early adopters report up to 60% time savings on tasks like budgeting and report writing. But the real power lies beyond automation - it's about freeing human potential for strategic thinking and innovation. As AI becomes the new standard, the challenge isn't job loss, but rapid adaptation. How is your organization preparing for this shift? #AIatWork #FutureOfWork #BusinessInnovation #DigitalTransformation #AIStrategy What's your experience with AI in the workplace? Share your thoughts below.

5