The most important news pieces from the AI world are here for you, whether you’re an AI enthusiast or just wish to keep an eye on what’s going on.
Focus Topic: Will AI take my job? OpenAI’s answer: Not if you get certified
“What’s AI going to mean for my job?” is quickly becoming the most asked and feared question in the workplace. While some roles are clearly at risk of automation, OpenAI’s latest move is sending a different signal: your job might not disappear – if you’re ready to evolve.
This week, OpenAI announced two major initiatives to expand economic opportunity through AI. First, it’s building a Jobs Platform to connect AI-skilled workers with companies that need them – backed by partners like Walmart, BCG, and state governments. Second, it’s launching OpenAI Certifications, which will validate AI fluency directly inside ChatGPT using a new “Study” mode. By 2030, OpenAI plans to certify 10 million Americans.
This isn’t about click-through courses or generic online diplomas. It’s about making sure AI skills are practical, visible, and hireable – for real companies solving real problems. Whether you’re an office worker, a local business owner, or someone changing careers, OpenAI’s new tools aim to reduce the friction between learning AI and getting paid for using it.
So, will you be out of work soon? Not if you stay ahead of the curve. The AI economy is already shifting from who can build the models to who knows how to use them effectively. And OpenAI just handed you a roadmap. The future may not be jobless – but it will definitely be skill-biased.
And that’s exactly why I care about raising AI awareness and why I train people in this field. I often warn startup founders: “Don’t just build a wrapper around ChatGPT – OpenAI or Google will do it better and faster, and you’ll be out of business.” But now? Maybe I’ll be out of work too… because ChatGPT is not only teaching AI but soon handing out certificates as well. 🙃
LLMs & AI Models
- Mistral AI upgraded its Le Chat platform for enterprise use with 20+ integrations and a new “Memories” function for persistent context.
- Tilde released an open-source large language model for smaller European languages, built to align with EU privacy rules and trained on supercomputers.
- DeepSeek is developing a self-improving AI agent that can learn from its past actions and complete multi-step tasks autonomously.
- Google launched EmbeddingGemma, an open-source model that runs offline on devices, processing over 100 languages in real time without internet.
- OpenAI made the ChatGPT projects feature available for free users, including memory, file uploads, icons, and customizable colors.
- Anthropic hit a $183B valuation after raising $13B, tripling its worth in 6 months, now serving over 300K business customers.
New Tools
- Lovable introduced a Voice Mode that lets users code and build apps via speech using ElevenLabsvoice recognition.
- NotebookLMadded new audio styles like “Debate”, “Critique”, and “Overview” to customize spoken summaries.
- Tencent launched HunyuanWorld-Voyager, a 3D model that creates navigable virtual worlds from a single photo.
- Google Photos integrated Veo 3, turning images into 4-second AI videos with better visuals – available only in the US for now.
👉 Explore these tools: Lovable | NotebookLM | Tencent Voyager | Veo 3
Other Quick Picks
- Anguilla earned $39M from .ai domain sales, now a quarter of the country’s total revenue.
- UCLA built an AI-based brain-computer interface to let paralyzed patients control robotic arms with their thoughts.Johannes Lee, Jonathan Kao, Neural Engineering and Computation Lab/UCLA
- Microsoft supports the US AI education push with free Copilot access for students and $1M in teacher grants.
- A US judge ruled Google can keep Chrome and Android, but must share data and drop exclusive deals.
- Apple plans to use Google Gemini to upgrade Siri into a multimodal assistant by 2026.
- Skepast&Puhkim cut engineering time and salary costs by 50% using ChatGPT and Python automation.
- Florida State University study shows AI phrasing is spreading into everyday speech and podcasts.
- MIT developed VaxSeer, an AI that predicts dominant flu strains better than WHO’s current system.
- xAI sued ex-engineer for stealing Grok model secrets before joining OpenAI.
- An AI stethoscope doubled heart failure detection and found arrhythmias 3.5x better than traditional tools.
- Meta’s AI lab is unstable – top talent leaving due to bureaucracy and low-quality data from Scale AI.
- Salesforce laid off 4,000 support staff as AI agents now handle 50% of customer conversations.
🇪🇪 AI News from Estonia
- Selver ad glitch showed an old man merged with vegetables – another case of sloppy AI design.
🎙️ AIPowerment Podcast dropped a new end-of-month episodefeaturing Sandra Roosna ( Askly ) sharing how to build an AI-powered sales assistant and why off-the-shelf bots won’t cut it.
🎧 Listen to AIPowerment Podcast on Spotify, Apple Podcasts, and YouTube.
Want to stay in the loop?
Straight to your inbox: practical AI updates, finance use cases, tools to try, and upcoming trainings.
I cut the noise and send only what’s actually useful.
💼 Exploring how to make AI work in finance? Let’s talk practical use cases – connect with Gerlyn Tiigemäe for expert guidance.
📚 If you would like to participate in one of my trainings or listen to speaking engagements, here are the upcoming ones:
- Training “AI võimalused finantsvaldkonnas” @ Äripäeva Akadeemia, 21.05.2025-28.11.2025 – information
- Training “AI võimalused projektijuhtimises” @ Äripäeva Akadeemia, 17.10.2025-28.11.2025 – information
