News

Republic World
republicworld.com > initiatives > sber-unveils-next-generation-ai-assistant-gigachat-based-on-flagship-model

Sber Unveils Next-Generation AI Assistant GigaChat Based on Flagship Model  

19+ hour, 22+ min ago  (404+ words) Sber has introduced a major update to its AI assistant GigaChat, which is now powered by the new flagship model GigaChat Ultra. This updated version allows the AI assistant to remember facts about users for personalized communication and solutions, search…...

Republic World
republicworld.com > tech > what-is-isara-openais-new-650m-love-interest-in-ai-agents

What Is Isara? OpenAI’s New $650M Love Interest in AI Agents

2+ day, 4+ hour ago  (270+ words) Isara is building AI agents that can work together, communicate, and solve complex problems. Unlike traditional AI models that operate in isolation, Isara's vision is to create swarms of agents that collaborate, almost like digital teams. Their early demonstrations have…...

Republic World
republicworld.com > initiatives > ai-has-learned-to-talk-now-it-has-to-remember

AI Has Learned to Talk. Now It Has to Remember

1+ week, 4+ day ago  (325+ words) Sam Altman has described current AI memory systems as "very crude, very early " GPT-2 of memory." Analyses of ChatGPT's architecture suggest its memory relies on lightweight layers: session metadata, explicitly saved facts, compressed summaries, and a sliding context window. This…...

Republic World
republicworld.com > viral > maryam-mirzakhani-hated-maths-to-fields-medal

From 'Hating Math' To Winning The Fields Medal: Story of Iran’s Genius Maryam Mirzakhani

1+ week, 6+ day ago  (406+ words) Updated 15 March 2026 at 13:37 IST Tehran: It is the ultimate underdog story in science. Before she was a world-renowned professor at Stanford University, or the first woman to win the Fields Medal (the "Nobel Prize of Math"). Maryam Mirzakhani was just…...

Republic World
republicworld.com > initiatives > privacy-first-voice-ai-why-klarna-chose-wispr-flow

Privacy-First Voice AI: Why Klarna Chose Wispr Flow

3+ week, 4+ day ago  (515+ words) As voice AI becomes more embedded in professional workflows, privacy compliance is increasingly influencing purchasing and deployment decisions. This is particularly relevant in Europe and the UK, where regulatory standards for data protection are among the most rigorous globally." The…...

Republic World
republicworld.com > initiatives > indian-origin-aws-specialist-recognized-for-advanced-role-in-cloud-infrastructure

Indian-Origin AWS Specialist Recognized for Advanced Role in Cloud Infrastructure

1+ mon, 3+ week ago  (435+ words) Hardeep Singh Tiwana, a Senior Technical Account Manager at Amazon Web Services (AWS), is among those advising major organizations on the transition to and operation of mission-critical Kubernetes environments. In the global cloud-native ecosystem, the stability of large-scale digital platforms…...

Republic World
republicworld.com > tech > open-source-ai-models-including-meta-s-llama-may-be-vulnerable-to-criminal-misuse-warn-researchers

Open-Source AI Models, Including Meta's Llama, May be Vulnerable to Criminal Misuse, Warn Researchers

1+ mon, 4+ week ago  (182+ words) While thousands of open-source LLM variants exist, a significant portion of the LLMs on the internet-accessible hosts are variants of Meta's Llama, Google DeepMind's Gemma, and others, according to the researchers. While some of the open-source models include guardrails, the…...

Republic World
republicworld.com > tech > can-ai-replace-humans-arnab-challenges-ai-s-structural-ceiling-on-originality-in-unscripted-showdown

Can AI Replace Humans? Arnab Unmasks AI's 'Structural Ceiling On Originality' In Unscripted Showdown

2+ mon, 2+ week ago  (236+ words) The conversation underscores why fears of AI "taking over" humans may be overstated, at least for now. According to the arguments presented, AI lacks agency, ambition, and the emotional or ethical stakes that drive human innovation. It does not rebel…...

Republic World
republicworld.com > tech > openai-confirms-gpt-5-hallucinations-heres-why-the-ai-gives-confidently-wrong-answers

OpenAI Confirms GPT-5 Hallucinations: Here’s Why the AI Gives Confidently Wrong Answers

6+ mon, 2+ week ago  (321+ words) When the AI boom took off, a new word quickly entered the public conversation - hallucinations. People were puzzled: how can software "hallucinate? Today, OpenAI itself is clarifying the picture. In a new research paper and blog post, the company admits…...

Republic World
republicworld.com > initiatives > how-to-use-the-margin-trading-facility-and-its-calculator

How to Use the Margin Trading Facility and Its Calculator?

8+ mon, 3+ week ago  (839+ words) Before diving into the calculator, it's crucial to understand'what is margin trading. In simple terms, it allows investors to buy shares by paying only a part of the total value. The remaining amount is financed by the broker. It is…...