DeepSeek-V3 Deprecated

# [AI Weekly News Rundown from April 13th to April 20th 2025](https://podcasts.apple.com/ca/podcast/ai-daily-news-rundown-april-18-2025-google-launches/id1684415169?i=1000704067566)

**Significant advancements and shifts characterised the AI landscape this past week.** Microsoft unveiled a highly energy-efficient AI model, while OpenAI's newest reasoning models surprisingly exhibited increased inaccuracies. Concerns arose among US chipmakers regarding Huawei potentially dominating China's AI market due to trade restrictions. In a display of technological progress, China hosted a half-marathon featuring humanoid robots. The business world saw Johnson & Johnson highlighting the concentrated value of specific AI applications, and an Italian newspaper experimented by letting AI manage an entire edition, revealing its ironic capabilities.

https://preview.redd.it/l0qlvyofhuve1.png?width=3000&format=png&auto=webp&s=9ecd0a9ca1f3b78559f23e2b116d7fbce4478380

**Like and Subscribe at** [**https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-deepseek/id1684415169**](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-deepseek/id1684415169)

🚀 **Djamgatech PRO**: Ad-Free Certification Mastery Unlock 50+ Professional Certifications with Premium Features

Why Upgrade to PRO? Zero Ads – Distraction-free studying Unlimited Tests – No daily limits Priority Support Advanced Analytics – Track performance

Djamgatech PRO [iOs](https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762): [https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762](https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762)

Djamgatech PRO [android](https://play.google.com/store/apps/details?id=com.coludeducation.quiz): [https://play.google.com/store/apps/details?id=com.coludeducation.quiz](https://play.google.com/store/apps/details?id=com.coludeducation.quiz)

Djamgatech FREE [Web/PWA](https://djamgatech.web.app/): [https://djamgatech.web.app](https://djamgatech.web.app)

# ⚡️ [Microsoft Researchers Create Super‑Efficient AI](https://arstechnica.com/ai/2025/04/microsoft-researchers-create-super%E2%80%91efficient-ai-that-uses-up-to-96-less-energy/)

* Microsoft has unveiled BitNet b1.58 2B4T, a "1-bit" AI model that operates on CPUs, including Apple's M2, using up to 96% less energy than traditional models. With 2 billion parameters trained on 4 trillion tokens, it matches the performance of larger systems while being more energy-efficient. \[[Read More](https://techcrunch.com/2025/04/16/microsoft-researchers-say-theyve-developed-a-hyper-efficient-ai-model-that-can-run-on-cpus/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* 🤔 [OpenAI’s New Reasoning AI Models Hallucinate More](https://techcrunch.com/2025/04/18/openais-new-reasoning-ai-models-hallucinate-more/)OpenAI's latest models, o3 and o4-mini, designed for enhanced reasoning, exhibit higher hallucination rates. Internal tests show o3 hallucinated 33% of the time on PersonQA, doubling the rate of its predecessor, o1. O4-mini performed worse, with a 48% hallucination rate. \[[Read More](https://mashable.com/article/openai-o3-o4-mini-hallucinate-higher-previous-models)\] \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* 💥 [Chipmakers Fear They Are Ceding China’s AI Market to Huawei](https://thebulletinbox.com/feed/6259044/us-chipmakers-fear-they-are-ceding-chinas-ai-market-to-huawei)U.S. chipmakers express concern over losing ground in China's AI market to Huawei, especially after new U.S. trade restrictions. Investigations are underway into potential export control violations, including Nvidia's alleged provision of restricted AI chips to Chinese firms. \[[Read More](https://www.huaweicentral.com/us-curbs-may-accelerate-huaweis-rise-in-chinese-ai-market-says-chip-firms/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* 🏃 [China Pits Humanoid Robots Against Humans in Half-Marathon](https://www.cnn.com/2025/04/19/world/video/robots-humans-half-marathon-beijing-digvid)In a world-first event, 21 humanoid robots competed alongside thousands of human runners in Beijing's Yizhuang half-marathon. The standout robot, Tiangong Ultra, completed the race in 2 hours and 40 minutes, showcasing China's advancements in robotics and AI. \[[Read More](https://talksport.com/sport/3133194/yizhuang-half-marathon-robots-china-beijing-athletics/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* 📊 [Johnson & Johnson: 15% of AI Use Cases Deliver 80% of Value](https://www.pymnts.com/news/artificial-intelligence/2025/johnson-15percent-ai-use-cases-deliver-80percent-value/)According to Johnson & Johnson’s global head of AI, just 15% of its AI initiatives generate 80% of its business value. These impactful use cases are often tied to supply chain optimization, manufacturing automation, and R&D acceleration. The company is refocusing its AI efforts on these high-yield domains. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* 📰 [Italian Newspaper Gives AI Free Rein—and Admires Its Irony](https://www.reuters.com/technology/artificial-intelligence/italian-newspaper-gives-free-rein-ai-admires-its-irony-2025-04-18/)An Italian newspaper handed over editorial duties to an AI assistant for a day, publishing an entire edition written and curated by the model. Editors were impressed by the AI's grasp of irony and nuanced commentary, though some warned of the potential for misinformation. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* 🧑‍💼 [AI-Powered Fake Job Seekers Are Flooding the Market](https://www.cbsnews.com/news/fake-job-seekers-flooding-market-artificial-intelligence/)Recruiters report a surge in applications from job seekers using AI-generated résumés, cover letters, and even voice avatars during interviews. Some applicants have even used AI-generated portfolios and fake work histories, complicating the hiring process and triggering new verification challenges.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/19](https://djamgatech.web.app/)\]
* ⚡️ [Google Launches Gemini 2.5 Flash with 'Thinking Budget'](https://developers.googleblog.com/en/start-building-with-gemini-25-flash/)Google has rolled out Gemini 2.5 Flash, a streamlined AI model designed for speed and cost-efficiency. A new feature, the "thinking budget," lets developers balance reasoning complexity with price and latency. This makes Gemini 2.5 Flash ideal for scalable, real-time applications like chatbots and content generation pipelines. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\]
* 📍 [Viral ChatGPT Trend: Reverse Location Searching from Photos](https://techcrunch.com/2025/04/17/the-latest-viral-chatgpt-trend-is-doing-reverse-location-search-from-photos/)A trending use of ChatGPT involves uploading travel photos and asking the AI to guess the location using visual clues alone—no metadata needed. This viral activity highlights how powerful multimodal models have become at interpreting geographical features, architecture, and landmarks. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\]
* 👀 [Meta Reportedly Asked Amazon and Microsoft to Fund Llama](https://www.linkedin.com/posts/theinformation_meta-asked-amazon-microsoft-to-help-fund-activity-7319034506034708482-_9ZR)In a bid to share the financial weight of training frontier models, Meta approached Amazon and Microsoft to help fund its open-source LLM project, Llama. While no formal deals have been confirmed, this suggests an evolving AI funding landscape, even among rivals.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\]
* ⚡️ [Google Launches Gemini 2.5 Flash with 'Thinking Budget'](https://developers.googleblog.com/en/start-building-with-gemini-25-flash/)Google has introduced Gemini 2.5 Flash, an AI model featuring a 'thinking budget' that allows developers to control the level of computational reasoning used for different tasks. This enables a balance between quality, cost, and response time, making the model adaptable for various applications. \[[Read More](https://developers.googleblog.com/en/start-building-with-gemini-25-flash/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🧬 [Profluent Discovers Scaling Laws for Protein-Design AI](https://fortune.com/2025/04/16/biotech-profluent-ai-scaling-laws-protein-design-models-opencrispr-openantibodies/)Biotech startup Profluent has identified 'scaling laws' in AI models used for protein design, indicating that larger models with more data yield predictably better results. This discovery enhances the potential for designing complex proteins, such as antibodies and genome editors, more effectively. \[[Read More](https://fortune.com/2025/04/16/biotech-profluent-ai-scaling-laws-protein-design-models-opencrispr-openantibodies/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 📊 [Transform Your Spreadsheets with AI in Google Sheets](https://workspace.google.com/resources/spreadsheet-ai/)Google Sheets now integrates AI capabilities through the 'Help me organize' feature, enabling users to create tables, structure data, and reduce errors efficiently. This enhancement aims to streamline data management and analysis within spreadsheets. \[[Read More](https://workspace.google.com/resources/spreadsheet-ai/)\] \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🤖 [Meta’s FAIR Shares New AI Perception Research](https://ai.meta.com/blog/meta-fair-updates-perception-localization-reasoning/)Meta's Fundamental AI Research (FAIR) team has released new research artifacts focusing on perception, localization, and reasoning. These advancements contribute to the development of more sophisticated AI systems capable of understanding and interacting with the environment. \[[Read More](https://ai.meta.com/blog/meta-fair-updates-perception-localization-reasoning/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🧠 [OpenAI Unveils o3 and o4-mini Reasoning Models](https://openai.com/index/introducing-o3-and-o4-mini/)OpenAI has released two new AI models: o3, its most advanced reasoning model to date, and o4-mini, a smaller, faster version optimized for efficiency. Both models can "think" with images, integrating visual inputs like sketches and whiteboards into their reasoning processes. They also have access to the full suite of ChatGPT tools, including web browsing, Python execution, and image generation. \[[Read More](https://openai.com/index/thinking-with-images/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 📱 [Perplexity AI to Be Pre-Installed on Motorola and Samsung Smartphones](https://www.androidcentral.com/apps-software/ai/perplexity-ai-motorola-samsung-agreement-discussions-reported)Perplexity AI is expanding its presence in the smartphone market by securing a deal with Motorola to preload its AI assistant on upcoming devices. The company is also in early talks with Samsung for potential integration. This move positions Perplexity as a competitor to established AI assistants like Google's Gemini. \[[Read More](https://m.economictimes.com/tech/artificial-intelligence/perplexity-ai-in-talks-to-integrate-assistant-into-samsung-motorola-phones/articleshow/120408775.cms)\] \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 💰 [OpenAI in Talks to Acquire Windsurf for $3 Billion](https://www.reuters.com/technology/artificial-intelligence/openai-talks-buy-windsurf-about-3-billion-bloomberg-news-reports-2025-04-16/)OpenAI is reportedly in advanced discussions to acquire Windsurf, an AI-powered coding assistant formerly known as Codeium, for approximately $3 billion. If finalized, this would be OpenAI's largest acquisition to date, potentially enhancing its capabilities in AI-assisted coding and intensifying competition with Microsoft's Copilot. \[[Read More](https://techcrunch.com/2025/04/17/openai-pursued-cursor-maker-before-entering-into-talks-to-buy-windsurf-for-3b/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🚫 [Meta Blocks Apple Intelligence Features on Its iOS Apps](https://www.pcmag.com/news/meta-blocks-apple-intelligence-features-on-facebook-instagram-threads)Meta has disabled Apple Intelligence features across its iOS applications, including Facebook, Instagram, Threads, Messenger, and WhatsApp. This move prevents users from accessing Apple's AI-powered tools like Writing Tools and Genmoji within these apps. \[[Read More](https://tech.slashdot.org/story/25/04/17/1539231/meta-blocks-apple-intelligence-in-ios-apps?utm_medium=feed&utm_source=rss1.0mainlinkuser)\] \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🖱️ [Copilot Gets Hands-On Computer Use](https://www.theverge.com/news/649574/microsoft-copilot-studio-computer-use-ai)Microsoft has introduced a new "computer use" feature in Copilot Studio, enabling AI agents to interact directly with websites and desktop applications. This allows the AI to perform tasks such as clicking buttons, selecting menus, and entering data into fields, effectively simulating human interaction with software that lacks API integrations. The feature is designed to adapt to changes in user interfaces, ensuring continued functionality even when buttons or screens are altered. \[[Read More](https://www.theverge.com/news/649574/microsoft-copilot-studio-computer-use-ai)\] \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🔒 [How to Run AI Privately on Your Own Computer](https://www.pcmag.com/how-to/how-to-run-your-own-chatgpt-like-llm-for-free-and-in-private)Running AI models locally ensures privacy and control over your data. Tools like GPT4All and Ollama allow users to operate AI chatbots on personal devices without internet connectivity. These applications are compatible with various operating systems and can run on standard hardware, making private AI accessible to a broader audience. \[[Read More](https://www.pcmag.com/how-to/how-to-run-your-own-chatgpt-like-llm-for-free-and-in-private)\] \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🧠 [Claude Gains Autonomous Research Powers](https://venturebeat.com/ai/claude-just-gained-superpowers-anthropics-ai-can-now-search-your-entire-google-workspace-without-you/)Anthropic's Claude AI assistant has been enhanced with a new "Research" feature, enabling it to autonomously search public websites and internal work resources to provide comprehensive answers. Additionally, integration with Google Workspace allows Claude to access data from Gmail, Docs, Sheets, and Calendar, improving its responsiveness and task efficiency. \[[Read More](https://venturebeat.com/ai/claude-just-gained-superpowers-anthropics-ai-can-now-search-your-entire-google-workspace-without-you/)\]\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 📚 [Wikipedia Offers AI Developers a Legit Dataset to Deter Bot Scrapers](https://www.theverge.com/news/650467/wikipedia-kaggle-partnership-ai-dataset-machine-learning)Wikipedia is collaborating with Kaggle to release a curated dataset for AI developers. The initiative aims to provide high-quality, structured data as an alternative to unauthorized bot scraping. The Wikimedia Foundation hopes this move will promote ethical AI development while reducing server strain from web crawlers.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🤖 [AI Support Agent Causes Uproar by Inventing Fake Policy](https://arstechnica.com/ai/2025/04/cursor-ai-support-bot-invents-fake-policy-and-triggers-user-uproar/)An AI assistant from Cursor, a coding-focused AI company, fabricated a policy during a user support interaction, causing confusion and backlash. The company has issued an apology, attributing the error to the model’s “hallucination” under high-volume use.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🎓 [Google One AI Premium Is Free for College Students Until Spring 2026](https://www.theverge.com/news/650921/google-one-ai-premium-gemini-free-college-education)Google is offering its $19.99/month Gemini AI Premium subscription for free to college students with verified .edu email addresses. The plan includes access to Gemini Advanced features like Gemini 1.5 Pro, Docs, Gmail integration, and AI-powered tools.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🧑‍💻 [New Technique Guides LLMs to Follow Programming Syntax More Reliably](https://news.mit.edu/2025/making-ai-generated-code-more-accurate-0418)MIT researchers have developed a method that steers large language models toward generating outputs that strictly adhere to syntax rules. The system doesn’t require retraining and uses model-agnostic prompting strategies to improve accuracy in code generation and data formatting. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/18](https://djamgatech.web.app/)\]
* 🔥 [The AI Chip War Burns Nvidia Again](https://www.reuters.com/markets/us/global-markets-view-usa-2025-04-16/)Nvidia is facing fresh setbacks as the U.S. government tightens export restrictions on AI chips destined for China. Specifically, the newly launched H20 chips—central to Nvidia’s AI roadmap—now require export licenses, effectively cutting off access to one of the company's largest markets. This development is expected to cost Nvidia an estimated $5.5 billion in revenue, according to analysts, and has triggered a 6% drop in its stock price.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 💥 [OpenAI Is Building a Social Network](https://www.theverge.com/openai/648130/openai-social-network-x-competitor)OpenAI is developing a social media platform that integrates ChatGPT's image generation into a social feed. This move aims to compete with Elon Musk's X (formerly Twitter) and gather user-generated data to enhance AI training. CEO Sam Altman has been seeking external feedback on the project, which is still in early stages. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 📉 [Nvidia Expects $5.5B Hit as US Targets Chips Sent to China](https://www.thetimes.co.uk/article/nvidia-faces-55bn-hit-from-trump-clampdown-on-ai-chips-qkl5d03nq)Nvidia anticipates a $5.5 billion financial impact due to new U.S. government restrictions on exporting its H20 AI chips to China. The measures aim to prevent these chips from supporting China's development of AI-powered supercomputers. The announcement led to a nearly 6% drop in Nvidia's shares in after-hours trading. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🗣️ [Anthropic Is Reportedly Launching a Voice AI You Can Speak To](https://www.theverge.com/news/649175/anthropic-claude-ai-voice-mode-report)Anthropic is preparing to introduce a "voice mode" feature for its Claude AI chatbot, offering three distinct voice options: Mellow, Airy, and Buttery. This feature aims to enhance user interaction by allowing more natural conversations with AI. The rollout is expected to begin as soon as this month. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🔮 [Grok Can Now Generate Documents, Code, and Browser Games](https://techcrunch.com/2025/04/15/grok-gains-a-canvas-like-tool-for-creating-docs-and-apps/)xAI's chatbot Grok has introduced "Grok Studio," a canvas-like tool that enables users to create and edit documents, code, and even browser-based games. The feature includes real-time collaboration and Google Drive integration, enhancing Grok's utility beyond simple chat interactions. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🎬 [Kling AI 2.0 Launches with Multimodal Video and Image Generation](https://www.globenewswire.com/news-release/2025/04/15/3062142/0/en/Kling-AI-Advances-to-the-2-0-Era-Empowering-Everyone-to-Tell-Great-Stories-with-AI.html)Kling AI has unveiled its 2.0 update, introducing a multimodal visual language (MVL) system that allows users to generate and edit videos and images using a combination of text, images, and video clips. The new version boasts significant improvements in motion quality, semantic responsiveness, and visual aesthetics, positioning it ahead of competitors like Google Veo2 and Runway Gen-4 in internal benchmarks.**What this means:** Kling AI 2.0's advancements in multimodal content creation empower users to produce high-quality, customized media, marking a significant step forward in AI-driven storytelling. \[[Watch](https://www.youtube.com/watch?v=a9KhKC04Hjo)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 📊 [Build a Personal AI Data Analyst with n8n Automation](https://n8n.io/workflows/3050-build-your-first-ai-data-analyst-chatbot/)n8n offers a workflow template that enables users to create an AI-powered data analyst chatbot. By connecting to data sources like Google Sheets or databases, the AI agent can perform calculations and deliver insights through platforms such as Gmail or Slack. This setup allows for efficient and automated data analysis without extensive coding knowledge. \[[Watch](https://www.youtube.com/watch?v=eScD4Y7nRBQ)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🕵️ [AI Models Play Detective in Ace Attorney](https://x.com/haoailab/status/1912231343372812508)Researchers at UC San Diego's Hao AI Lab tested leading AI models on their ability to play the game Phoenix Wright: Ace Attorney. The AI agents were tasked with identifying contradictions and presenting evidence in court scenarios. While models like OpenAI's GPT-4.1 and Google's Gemini 2.5 Pro showed some success, none fully solved the cases, highlighting the challenges AI faces in complex reasoning tasks. \[[2025/04/16](https://djamgatech.web.app/)\]
* 🏛️ [Trump’s AI Infrastructure Plans May Be Delayed by Texas Republicans](https://www.theguardian.com/us-news/2025/apr/15/trump-texas-ai-infrastructure-republicans)Former President Donald Trump’s ambitious plans to build a national AI infrastructure could face opposition from members of his own party in Texas. Some state Republicans are resisting federal AI development initiatives, citing concerns about data privacy, government overreach, and unclear economic benefits. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🔊 [Humans Struggle to Identify AI-Generated Deepfake Voices](https://www.newscientist.com/article/2476556-people-are-really-bad-at-spotting-ai-generated-deepfake-voices/)A new study published in \*New Scientist\* shows that people consistently fail to distinguish AI-generated deepfake voices from real ones. Even experienced listeners were wrong more than half the time, raising alarm about how easily synthetic audio can be used to deceive.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🤖 [Hugging Face Acquires Humanoid Robotics Startup](https://techcrunch.com/2025/04/14/hugging-face-buys-a-humanoid-robotics-startup/)Hugging Face has acquired an unnamed humanoid robotics company to expand its portfolio beyond large language models. The move signals Hugging Face’s ambitions to integrate AI models into embodied agents that can interact with the physical world.**What this means:** This acquisition hints at a future where open-source AI tools are increasingly embedded into real-world robotics, potentially accelerating development in autonomous systems and personal robotics. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🖼️ [ChatGPT Adds Personal Image Library for AI-Generated Art](https://www.theverge.com/news/649247/chatgpt-image-library)OpenAI has introduced a new "image library" section in ChatGPT, allowing users to view and manage all their AI-generated images. The feature enhances accessibility and user control over creative assets, and it works across both desktop and mobile platforms. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🧠 [OpenAI Debuts GPT-4.1 Flagship AI Model](https://www.theverge.com/news/647896/openai-chatgpt-gpt-4-1-mini-nano-launch-availability)OpenAI has released GPT-4.1, its latest flagship AI model, featuring significant enhancements in coding, instruction following, and long-context comprehension. The model supports up to 1 million tokens, a substantial increase from previous versions. GPT-4.1 is available in three variants: the standard model, a cost-effective Mini version, and a lightweight Nano version, which is the fastest and most affordable to date. \[[OpenAI Announcement](https://openai.com/index/gpt-4-1/)\] \[[Reuters Coverage](https://www.reuters.com/technology/artificial-intelligence/openai-launches-new-gpt-41-models-with-improved-coding-long-context-2025-04-14/)\] \[[Wired Analysis](https://www.wired.com/story/openai-announces-4-1-ai-model-coding)\]
* 👀 [Apple Plans to Improve AI Models by Privately Analyzing User Data](https://www.bloomberg.com/news/articles/2025-04-14/apple-to-analyze-user-data-on-devices-to-bolster-ai-technology)Apple is set to enhance its AI capabilities by analyzing user data directly on devices, ensuring that personal information remains private. This approach leverages techniques like differential privacy and synthetic data generation to train AI models without compromising user confidentiality.\[[Business Insider Report](https://www.businessinsider.com/apple-intelligence-ai-training-privacy-synthetic-data-2025-4)\] \[[ZDNet Explanation](https://www.zdnet.com/article/how-apple-plans-to-train-its-ai-on-your-data-without-sacrificing-your-privacy/)\] \[[Yahoo Finance Article](https://finance.yahoo.com/news/apple-details-plans-improve-ai-105355573.html)\]
* 🫠 ["Slopsquatting" Attacks Exploit AI-Hallucinated Package Names](https://www.bleepingcomputer.com/news/security/ai-hallucinated-code-dependencies-become-new-supply-chain-risk/)A new cybersecurity threat known as "slopsquatting" has emerged, where attackers register fake package names that AI models mistakenly suggest during code generation. Developers who unknowingly use these hallucinated package names may introduce malicious code into their software projects.\[[Infosecurity Magazine Insight](https://www.infosecurity-magazine.com/news/ai-hallucinations-slopsquatting/)\] \[[The Register Coverage](https://www.theregister.com/2025/04/12/ai_code_suggestions_sabotage_supply_chain/)\] \[[Wikipedia Overview](https://en.wikipedia.org/wiki/Slopsquatting)\]
* 🎬 [ByteDance's Seaweed-7B: A Compact Powerhouse in AI Video Generation](https://seaweed.video/)ByteDance has introduced Seaweed-7B, a 7-billion-parameter diffusion transformer model designed for efficient video generation. Trained using 665,000 H100 GPU hours, Seaweed-7B delivers high-quality videos from text prompts or images, supporting resolutions up to 1280x720 at 24 FPS. Its capabilities include text-to-video, image-to-video, and audio-driven synthesis, making it a versatile tool for creators.\[[Read the Paper](https://arxiv.org/abs/2504.08685)\] \[[Watch Demo](https://www.youtube.com/watch?v=NdELnNBDLmo)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🧠 [Google's DolphinGemma: Decoding Dolphin Communication with AI](https://blog.google/technology/ai/dolphingemma/)Google, in collaboration with the Wild Dolphin Project and Georgia Tech, has developed DolphinGemma, an AI model trained on dolphin vocalizations. Utilizing Google Pixel phones, researchers aim to analyze and predict dolphin sounds, potentially enabling two-way communication through the CHAT system. \[[TechCrunch Coverage](https://techcrunch.com/2025/04/14/googles-newest-ai-model-is-designed-to-help-study-dolphin-speech/)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🧭 [Build a Personal AI Data Analyst with n8n Automation](https://n8n.io/workflows/3050-build-your-first-ai-data-analyst-chatbot/)n8n offers a workflow template that enables users to create an AI-powered data analyst chatbot. By connecting to data sources like Google Sheets or databases, the AI agent can perform calculations and deliver insights through platforms such as Gmail or Slack. This setup allows for efficient and automated data analysis without extensive coding knowledge. \[[Watch Tutorial](https://www.youtube.com/watch?v=eScD4Y7nRBQ)\] \[[2025/04/16](https://djamgatech.web.app/)\]
* 🚀 [Ilya Sutskever’s SSI Raises $2B at $32B Valuation](https://www.reuters.com/technology/artificial-intelligence/alphabet-nvidia-invest-openai-co-founder-sutskevers-ssi-source-says-2025-04-12/) Safe Superintelligence Inc. (SSI), co-founded by former OpenAI chief scientist Ilya Sutskever, has raised $2 billion in funding, bringing its valuation to $32 billion. The funding round was led by Greenoaks Capital, with participation from Alphabet and Nvidia. SSI is focused on developing a safe superintelligence, aiming to surpass human-level AI while ensuring safety remains paramount. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🧪 [AI Surpasses Experts in Tuberculosis Diagnosis](https://www.auntminnie.com/clinical-news/ultrasound/article/15742661/aiguided-pocus-bests-experts-in-diagnosing-tb-in-underserved-areas)Researchers at the ESCMID Global 2025 conference presented findings that an AI-guided point-of-care ultrasound (POCUS) system outperformed human experts by 9% in diagnosing pulmonary tuberculosis (TB). The AI model, ULTR-AI, achieved a sensitivity of 93% and specificity of 81%, exceeding WHO's target thresholds for non-sputum-based TB triage tests.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 📣 [Ex-OpenAI Staff Push Back on For-Profit Shift](https://www.therundown.ai/p/ilyas-ssi-skyrockets-to-32b)A group of former OpenAI employees have expressed concerns over the company's transition to a for-profit model. They argue that this shift undermines OpenAI's original mission to develop AI for the benefit of humanity and could compromise safety and ethical standards.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🤖 [Build an AI-Powered Lead Outreach Automation](https://www.therundown.ai/p/ilyas-ssi-skyrockets-to-32b)Developers and marketers are increasingly leveraging AI to automate lead outreach processes. By integrating AI models with tools like Zapier, businesses can create systems that automatically qualify leads, personalize communication, and streamline sales workflows. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🇺🇸 [Nvidia to Manufacture AI Supercomputers in the U.S.](https://www.investors.com/news/technology/nvidia-stock-us-made-ai-supercomputers/)Nvidia has announced plans to build AI supercomputers entirely within the United States, investing up to $500 billion over the next four years. The initiative includes producing Blackwell chips in Arizona and establishing supercomputer manufacturing plants in Texas, in collaboration with partners like TSMC, Foxconn, and Wistron. This move aims to strengthen supply chains and meet the growing demand for AI infrastructure. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🐬 [Google Develops AI Model to Decode Dolphin Communication](https://blog.google/technology/ai/dolphingemma/)Google has introduced DolphinGemma, an AI model designed to analyze and interpret dolphin vocalizations. Trained on decades of data from the Wild Dolphin Project, DolphinGemma can identify patterns in dolphin sounds and even generate dolphin-like sequences. The model runs efficiently on Pixel smartphones, facilitating real-time analysis in the field.\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🎨 [AI-Generated Action Figures Flood Social Media—Then Artists Reclaimed the Trend](https://www.nbcnews.com/tech/social-media/ai-action-figures-social-media-artists-hand-drawn-rcna201056) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🚀 [Google and Nvidia Invest in Ilya Sutskever’s Safe Superintelligence](https://www.businesstoday.in/technology/news/story/google-nvidia-invest-in-openai-co-founder-ilya-sutskevers-ai-startup-safe-superintelligence-471877-2025-04-14) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🗂️ [DeepSeek-V3 Deprecated on GitHub](https://github.blog/changelog/2025-04-11-deepseek-v3-is-now-deprecated-in-github-models/) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🪐 [High School Student Uses AI to Discover 1.5 Million Unknown Space Objects](https://phys.org/news/2025-04-high-school-student-ai-reveal.html) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]
* 🤖 [OpenAI’s Next AI Agent: A Self-Testing Software Engineer](https://www.livemint.com/technology/tech-news/openais-next-ai-agent-is-a-self-testing-software-engineer-that-does-what-humans-won-t-chatgpt-11744506780340.html) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/13](https://djamgatech.web.app/)\]
* 🎭 [‘Wizard of Oz’ AI Makeover Sparks Mixed Reactions](https://www.foxnews.com/entertainment/wizard-of-oz-ai-makeover-total-transformation-sparking-mixed-reactions-experts) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/13](https://djamgatech.web.app/)\]
* 💼 [Amazon CEO Lays Out AI Vision in Shareholder Letter](https://www.reuters.com/technology/amazon-ceo-sets-out-ai-investment-mission-annual-shareholder-letter-2025-04-10/)\[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/13](https://djamgatech.web.app/)\]
* 🎬 [James Cameron: Use AI to Cut Film Costs—But Keep the Crew](https://www.indiewire.com/news/general-news/james-cameron-ai-cut-cost-of-making-films-1235115173/) \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/13](https://djamgatech.web.app/)\]

https://preview.redd.it/fibknpydjuve1.png?width=1024&format=png&auto=webp&s=6a62335424e31fa8034945e98eb3beae8051d2f7

🚀 [Djamgatech PRO](https://djamgatech.web.app): Ad-Free Certification Mastery Unlock 50+ Professional Certifications with Premium Features

Djamgatech PRO [iOs](https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762): [https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762](https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762)

# [A Daily Chronicle of AI Innovations on April 14th 2025](https://podcasts.apple.com/ca/podcast/ai-weekly-news-rundown-april-07-to-april-13-2025-netflix/id1684415169?i=1000703307778)

https://preview.redd.it/gk85pj5pexue1.png?width=3000&format=png&auto=webp&s=695b6c6e2e435ffa86d7763d97fb165ac045590a

# 🚀 [Ilya Sutskever’s SSI Raises $2B at $32B Valuation](https://www.reuters.com/technology/artificial-intelligence/alphabet-nvidia-invest-openai-co-founder-sutskevers-ssi-source-says-2025-04-12/)

Safe Superintelligence Inc. (SSI), co-founded by former OpenAI chief scientist Ilya Sutskever, has raised $2 billion in funding, bringing its valuation to $32 billion. The funding round was led by Greenoaks Capital, with participation from Alphabet and Nvidia. SSI is focused on developing a safe superintelligence, aiming to surpass human-level AI while ensuring safety remains paramount.

* The [brief](https://link.mail.beehiiv.com/ss/c/u001.7FjCl1Hhb45GEizGv1NNbJuA4PP5jxEGVVB1XjKjm1Ob6u-oJ4rmwQYVQLXYhJWak0NnaMmBFOboqrJY5mIem-tbqBEmxrlFJkqwAM4Edi4nABjeY3FXwwi8T9mQh-W3nTT8SOPihWudcUWPOrAGwTJOGW5_6GS6Q67P9Q1sGNK3JB8ZhSgwOIFgYPUttMqd9k4clJ0bHhbWuKlJDS6PGw9Ao7sD0qjVf_Yn6LOrt0paEZfwL1WZt4oRYXQM5y8I4eo_Gh_0P8LRWLup9Crzvn6ZOcIjU-VukDwxOHKLCuzMcXd9Z95kg72GpRBWccdgSiOIaBCB-PyNU9hJ7kSLSQ/4fn/qBrIk1myTLCsZZag9JC8-A/h14/h001.3Vz1SHuZgzFSgs9z9PzNMh-JW2xKQS5_cNW72E-An8I) makes the case that if OpenAI’s non-profit wing cedes its controlling stake in business, it would “fundamentally violate its mission statement.”
* It adds that OpenAI’s restructuring would also “breach the trust of employees, donors, and other stakeholders” who supported the lab for its mission.
* Todor Markov, who is now at Anthropic, called Altman “a person of low integrity” who used the charter merely as a “smoke screen” to attract talent.
* They all noted the court should recognize maintaining the nonprofit is essential to ensure AGI benefits humanity rather than serving narrow financial interests.”

**What this means:** SSI's rapid ascent underscores investor confidence in Sutskever's vision for safe superintelligence, highlighting the growing emphasis on AI safety in the industry. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🧪 [AI Surpasses Experts in Tuberculosis Diagnosis](https://www.auntminnie.com/clinical-news/ultrasound/article/15742661/aiguided-pocus-bests-experts-in-diagnosing-tb-in-underserved-areas)

https://preview.redd.it/95apgsfsexue1.png?width=1292&format=png&auto=webp&s=b5799817dd6da588fbce232d5e70588271f542f4

Researchers at the ESCMID Global 2025 conference presented findings that an AI-guided point-of-care ultrasound (POCUS) system outperformed human experts by 9% in diagnosing pulmonary tuberculosis (TB). The AI model, ULTR-AI, achieved a sensitivity of 93% and specificity of 81%, exceeding WHO's target thresholds for non-sputum-based TB triage tests.

* Presented at ESCMID Global 2025, the study introduced ULTR-AI, an AI system trained to read lung ultrasound images from smartphone-connected devices.
* The system uses a combination of three different models to merge image interpretation and pattern detection and optimize diagnosis accuracy.
* When tested on 504 patients (38% of whom had confirmed TB), it achieved 93% sensitivity and 81% specificity, beating human expert performance by 9%.
* The AI can identify subtle patterns that humans often miss, including small pleural lesions invisible to the naked eye.

**What this means:** AI-powered diagnostic tools like ULTR-AI can enhance TB detection, especially in underserved areas, offering rapid, accurate, and non-invasive screening methods. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 📣 [Ex-OpenAI Staff Push Back on For-Profit Shift](https://www.therundown.ai/p/ilyas-ssi-skyrockets-to-32b)

A group of former OpenAI employees have expressed concerns over the company's transition to a for-profit model. They argue that this shift undermines OpenAI's original mission to develop AI for the benefit of humanity and could compromise safety and ethical standards.

**What this means:** The debate highlights the tension between commercial interests and ethical considerations in AI development, emphasizing the need for transparency and accountability. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🤖 [Build an AI-Powered Lead Outreach Automation](https://www.therundown.ai/p/ilyas-ssi-skyrockets-to-32b)

Developers and marketers are increasingly leveraging AI to automate lead outreach processes. By integrating AI models with tools like Zapier, businesses can create systems that automatically qualify leads, personalize communication, and streamline sales workflows.

1. Set your [Lindy AI](https://link.mail.beehiiv.com/ss/c/u001.dSnm3kaGd0BkNqLYPjeMf6Rj55ZWgZvEaP0t0oFKnl9Zsf5NqxzDC5aWyXuX9FCb8fdC6CiP6T7sJ-3UKFQXwJsXZqxzKfhSn4OsXOlxZQ0NS_pIpHwrOf2Vui8bvd5qVuueH4RRVvpS633RRytwf_OLI8dW1ocFIEibtzFf2aNywihcb1TANRmXKvJsoOCmxXZFwLGtKmsLPDKd-1SQ6SV_l6LTxfDfmtoT8x1Gugc/4fn/qBrIk1myTLCsZZag9JC8-A/h16/h001.ZS5X9KakJeTDDP_Btw7mWHXuGpvm6hp_3X8hTZaa1Fs) agent context by adding a description like "You are an outreach agent that has access to spreadsheets, researches leads, and drafts personalized emails".
2. Create a workflow starting with "Message Received" trigger and an AI Agent configured to process spreadsheets of leads.
3. Add an "Enter Loop" node that processes leads in parallel, with "Search Perplexity" and "Draft Email" nodes inside the loop.
4. Finalize with an "Exit Loop" node and a summary AI Agent, then test your workflow with a sample spreadsheet.

**What this means:** AI-driven automation can enhance efficiency in lead generation and outreach, allowing businesses to scale their operations and improve customer engagement. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🇺🇸 [Nvidia to Manufacture AI Supercomputers in the U.S.](https://www.investors.com/news/technology/nvidia-stock-us-made-ai-supercomputers/)

Nvidia has announced plans to build AI supercomputers entirely within the United States, investing up to $500 billion over the next four years. The initiative includes producing Blackwell chips in Arizona and establishing supercomputer manufacturing plants in Texas, in collaboration with partners like TSMC, Foxconn, and Wistron. This move aims to strengthen supply chains and meet the growing demand for AI infrastructure.

* Nvidia plans to manufacture AI supercomputers entirely in the U.S. for the first time, commissioning over a million square feet of manufacturing space in Arizona and Texas with partners like TSMC, Foxconn, and Wistron.
* The company aims to produce up to half a trillion dollars of AI infrastructure in the United States within the next four years through collaborations with global manufacturing leaders to strengthen supply chain resilience.
* Jensen Huang, Nvidia's CEO, emphasized that building AI chips and supercomputers in America will help meet growing demand, create hundreds of thousands of jobs, and drive trillions in economic security.

**What this means:** By localizing production, Nvidia seeks to enhance supply chain resilience and position itself at the forefront of AI development amid global trade tensions. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🐬 [Google Develops AI Model to Decode Dolphin Communication](https://blog.google/technology/ai/dolphingemma/)

Google has introduced DolphinGemma, an AI model designed to analyze and interpret dolphin vocalizations. Trained on decades of data from the Wild Dolphin Project, DolphinGemma can identify patterns in dolphin sounds and even generate dolphin-like sequences. The model runs efficiently on Pixel smartphones, facilitating real-time analysis in the field.

* Google has partnered with the Wild Dolphin Project to develop DolphinGemma, an AI model based on its Gemma framework that analyzes complex dolphin vocalizations and communication patterns.
* Researchers have already identified some dolphin sounds like signature whistles used as names and "squawk" patterns during fights, but they hope this AI collaboration will reveal if dolphins have a structured language.
* The new AI model uses Google's SoundStream technology to tokenize dolphin sounds, allowing real-time analysis of the marine mammals' complex whistles and clicks that have puzzled scientists for decades.

**What this means:** This advancement could pave the way for meaningful interspecies communication, offering insights into dolphin behavior and cognition. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🎨 [AI-Generated Action Figures Flood Social Media—Then Artists Reclaimed the Trend](https://www.nbcnews.com/tech/social-media/ai-action-figures-social-media-artists-hand-drawn-rcna201056)

AI-generated action figure portraits took social media by storm, depicting stylized versions of people as heroic characters. But soon, hand-drawn alternatives by traditional artists began trending as a counter-movement. Artists reclaimed the medium, offering more personal, expressive, and human-centered designs.

**What this means:** This cultural clash illustrates the ongoing dialogue between AI-generated content and human creativity, raising questions about authenticity and the value of hand-crafted art in the digital era. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🚀 [Google and Nvidia Invest in Ilya Sutskever’s Safe Superintelligence](https://www.businesstoday.in/technology/news/story/google-nvidia-invest-in-openai-co-founder-ilya-sutskevers-ai-startup-safe-superintelligence-471877-2025-04-14)

Safe Superintelligence (SSI), the AI startup co-founded by OpenAI’s former chief scientist Ilya Sutskever, has secured major backing from Google and Nvidia. The firm is focused on safely building AI systems that exceed human intelligence while staying aligned with human goals.

**What this means:** With leading tech giants backing SSI, the startup could become a key player in the global race to develop AGI—placing safety and alignment at the forefront. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🗂️ [DeepSeek-V3 Deprecated on GitHub](https://github.blog/changelog/2025-04-11-deepseek-v3-is-now-deprecated-in-github-models/)

GitHub has officially deprecated the DeepSeek-V3 model from its Models platform as of April 11. Developers are encouraged to migrate to newer, actively maintained alternatives. The deprecation follows the release of improved open-source models across the AI community.

**What this means:** The fast-paced evolution of open-source AI models is leading to shorter lifespans for legacy systems, pushing developers to stay updated with cutting-edge releases. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# 🪐 [High School Student Uses AI to Discover 1.5 Million Unknown Space Objects](https://phys.org/news/2025-04-high-school-student-ai-reveal.html)

A high school student has used AI algorithms to identify more than 1.5 million previously unclassified objects in space, using publicly available astronomical data. The discovery is hailed as one of the largest amateur contributions to modern astronomy.

**What this means:** AI democratizes discovery, enabling individuals—even students—to contribute meaningfully to scientific advancement with limited resources. \[[Listen](https://podcasts.apple.com/ca/podcast/ai-unraveled-latest-ai-news-trends-chatgpt-gemini-gen/id1684415169)\] \[[2025/04/14](https://djamgatech.web.app/)\]

# What Else Happened in AI on April 14th 2025?

**Meta’s** unmodified, release version of Llama 4 Maverick [appeared](https://link.mail.beehiiv.com/ss/c/u001.6k0_SAz8nrOuu_-LoNX1Hc37nxRkyMbMXhsZdbQrel1BcXqGnjgPiIcFHyeHy6JEU9_99puqTfNkgeQmW2mEz6WbFoi1-Gtz7Brfbe3U1klvrYlHBzqGD4kdQ3JD4BaR970Rc_CqJ31WOGF07SCimgIcJdZ_Yh9y0D-FB7hl4-cM6FoIY9oNrsF3RxrQ56-xfnZkDjnKllCZvIz62c9_Ne2-MJu0IDiytgCj8LyECxc5TPdK2ocinbW_jWVC50UIEADemZhK6CSCvVG72HqCAA/4fn/qBrIk1myTLCsZZag9JC8-A/h34/h001.xfi44IErNGVwGeOQ3T1N8E93xA9J9whDD7G4DJau4JU) on LMArena, ranking below months-old models, including Gemini 1.5 Pro and Claude 3.5 Sonnet.

**DeepMind** **CEO Demis Hassabis** [mentioned](https://link.mail.beehiiv.com/ss/c/u001.KT4rQsO6sHS_v2VASG2xukYrcBLmr-VWvDqpbYLTfcSWU8iVzktVnd5dk66ackEs8fq5FSzSpdFc1q2YFOzFPRK1iGXM-VkHxrGjCiMqVCilQDAUJsmKe-RfH0qJrxCMoQjMyr6Z7PcMrw_AFloKzbNrNyz4mrSxFfgUqxof0rgqxMgwGAHf3QSeUPY1nfo7TaVwNfcSCeKkaohHGT57DIPs1B14V8fnwSPTmmR3YUokmVyFelsuxA3Ol0jZ_UXimPD_FLdkcw66QWiQ6gmi5t_PSDebeqeLu8wmo9Ji4-GCNed4l94bdfvv_ZL6abu_c01qnBzayaCvo9O5mulC9KO6tAtMh8f65-jmKRarj_V-6AUjDEueJXHeE-3fm4Rq/4fn/qBrIk1myTLCsZZag9JC8-A/h35/h001._YhMTMgtFwVdN5MMCyhjk7zp4-73FgRrUkf8R9q4JQY) that the company plans to combine Gemini and Veo models into a unified omni model with better world understanding.

**Netflix** is [reportedly](https://link.mail.beehiiv.com/ss/c/u001.dSnm3kaGd0BkNqLYPjeMf_MnrMNPlyZa0tC_fQ34TxQ78dU03jIigb7dPAeatjyNw92IoJAp-RWqnDbkOTempIDdJkM4V1pdeFm-IU8gSTFQw98Yt4iEQ6vHfdazseGtz9gTWnOL25bxsfBVyjIobZ132w_Df2Y65fsLeAUvA1YyQAA9aOoEdU-6oSsmbdwDo7M6THws3knuAGdaMgvVdUXvxUrNV_3dR6KgZp2_2M7USYUxgUzI818kkrHNzrKr-uF4yUfv8_vgrNdqJ-kmvqG7ajrP9uMKDgLbDxgURHPHArRb2B4AEnexTz1wUF9EvSKfx5p_DBp1pUURsa2YvzXiKFqSDyzlwrNyXGbCUdk/4fn/qBrIk1myTLCsZZag9JC8-A/h36/h001.M9zcc7oT3gojg8jd3SZ5ksciPhverwnHz9MNAKIW_nY) working with OpenAI on a revamped search experience, allowing users to look up content using different new parameters, including their mood.

**OpenAI** [beefed](https://link.mail.beehiiv.com/ss/c/u001.ZY5Y0CT8KZaZ1y9TVLsmf4NuYOh8P8a4yE31n--08b8O2ARVJFGYxSVqZWMtAT-RKxdeDPsxinOrJgeC_XsVpPd6g4hgFlJYqhhQ_QnWytzhdXUpqNjhlTsWKghN3QBI5EsIS4W9YMzq2efDITNPXQyY7G9Wv1lkfJYjMcsSgxzhW3KFoMbO6XJxT1HVirLs9RPUp8MbOTZZmclwAV5cMzWQITWX9oiJ6tHZn3ks8vmLdWs3ybMBcSPHhaYlswhOrHsx26l2PLZbDhqo6qf6b5HGwRtSTNW0FMrFXxAQl88/4fn/qBrIk1myTLCsZZag9JC8-A/h37/h001.EwJcYTLlJ7atxyrf4YQ8j1r_8xJq_T6bhFw_VTbK2kI) up its security with a new Verified Organization status, which will be required to unlock API access to its advanced models and capabilities.

**OpenAI CEO Sam Altman** [said](https://link.mail.beehiiv.com/ss/c/u001.6k0_SAz8nrOuu_-LoNX1Hat-S5-YShOXuDICLVQQU-ufmx5Ln0eD80xMr8lW0gOWph-WtL6dAImfMgRGVwe7M3zj_mXSBAEfE8Mp5_km0muYSv5tJA-WqOsONzcSXtHePuCpyS0Ov1xt-zO0Au0xXZ_XcWXmZH04lggjZ5uW3Wnt3j0T4R8jkzM0_kMBz9MWQDqbfjHnJ6JwEXkO8uagsJ9AIGhbMFYpBuGo3BT9SJ0qmpoM4tenWA6b7Ne23v7Vspy01uwjw5xdOB_bYBgavA/4fn/qBrIk1myTLCsZZag9JC8-A/h38/h001.Vaol3qxOYzpIUcI2pcMIZOw5gZy2_yKLAkRJRDO8NXU) that the company plans to release an open-source model that would be "near the frontier.”

**Elon Musk’s xAI** started [rolling out](https://link.mail.beehiiv.com/ss/c/u001.6k0_SAz8nrOuu_-LoNX1HWwKk4Xf6pY4U2w5nPBMcy0VHjJ9CBBFbaQn8xcMDI-TlkwTu_XLvEibee5rVo2qTlLoO9PKGxiHacKPPk4e9mvzbBkbsbB4TXnvPijy9XNT-ldZfmFBm3FcooOD46NUt4H-IONRv4FbJG5DljlOKEoiYVL3QSxEMhb3vxSa9xe2wye7SJUd1NJlBLHF0z9PhnqRmS6i8NEpnykSi8CJ_iV1kkx3r1ISkJVuo1NqEMqWgTN1_VAyrX4XcYy-T2Qksw/4fn/qBrIk1myTLCsZZag9JC8-A/h39/h001.8ZLgTlYFd4Vc6CR0fJzVyq3DoBBoIOKDD3UxkbMwrnU) the memory feature to its Grok AI assistant, following a similar move from OpenAI last week.

#  🚀 Djamgatech PRO: Ad-Free Certification Mastery

https://preview.redd.it/8yefvvfvexue1.png?width=1024&format=png&auto=webp&s=576a6b5a65ce7e20a52bb1948d28d5b00ec245fa

Unlock 50+ Professional Certifications with Premium Features

Why Upgrade to PRO?

Zero Ads – Distraction-free studying

Unlimited Tests – No daily limits

Priority Support

Advanced Analytics – Track performance

EXCLUSIVE PRO FEATURES

1. AI-Powered Study Assistant
2. Curated Study Paths
3. Premium Question Bank

* 15,000+ questions (vs. 5,000 in free version)
* Includes simulated PBQs (AWS, Cisco, PMP)

CERTIFICATION COVERAGE

Cloud & DevOps AWS (12 certs) | Google Cloud (9) | Azure (7) | Kubernetes (CKA, CKAD)

Cybersecurity CISSP & Concentrations | OSCP | CEH | CCSP | CISM

Project Management PMP | PgMP | PMI-ACP | PRINCE2 | Scrum Master

Finance & Law CPA | CFA | FRM | CIPP/E | Patent Bar

Healthcare IT CPC | CCS | RHIA | Epic Certifications

PRO BENEFITS

✓ Download All Content – Study completely offline

✓ Unlimited Custom Quizzes – Focus on exact weak areas

✓ Early Access – New features 2 weeks before free tier

✓ Certificate Generator – Validate skills for employers

USER RESULTS

"PRO's PBQs helped me pass AWS Solutions Architect in 3 weeks!" ★★★★★ "

Worth every penny – the 90-day PMP plan got me certified first try." ★★★★★

"Finally, an app with CIPP/E materials!" ★★★★★

SPECIAL LAUNCH OFFER

7-Day Free Trial – Cancel anytime

Download Djamgatech PRO Today – Your Career Upgrade Starts Here!

Djamgatech FREE [Web/PWA](https://djamgatech.web.app/): [https://djamgatech.web.app](https://djamgatech.web.app)

Djamgatech PRO [iOs](https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762): [https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762](https://apps.apple.com/ca/app/djamgatech-certmaster-ai-pro/id1574297762)

Djamgatech PRO [android](https://play.google.com/store/apps/details?id=com.coludeducation.quiz): [https://play.google.com/store/apps/details?id=com.coludeducation.quiz](https://play.google.com/store/apps/details?id=com.coludeducation.quiz)

Djamgatech PRO [Windows](https://apps.microsoft.com/detail/9nnr153jfgw2?hl=en-gb&gl=CA): [https://apps.microsoft.com/detail/9nnr153jfgw2?hl=en-gb&gl=CA](https://apps.microsoft.com/detail/9nnr153jfgw2?hl=en-gb&gl=CA)