RE: https://mastodon.art/@otterlove/116505289006825916
This year's #StateOfAI survey had the question "What other questions do you wish the survey would've asked, if any?"
Here are my suggestions—I thought, I didn't want to keep them from you:
– How do you feel due to the fact that you are surrounded by #AI in your job and daily life?
1/
#amCoding #amProgramming #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #noAI #LLM
RE: https://mastodon.art/@otterlove/116505289006825916
If you don't like #AI, are an AI skeptic, or a #noAI supporter, you can participate as well and share your side of the story
#webDev #webDevelopment #softwareDevelopment #softwareEngineering #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #StateOfAI #StateOfWebDev #FollowerPower #LLM
The "honeymoon phase" of Generative AI is officially over. According to McKinsey’s latest state of AI report, adoption has spiked, with 65% of organizations now using GenAI regularly. However, the gap between the "experimenters" and the "achievers" is widening into a canyon. We’ve spent two years marveling at what the models can say; now we are finally being judged on what they can actually do for the P&L.
The insight here is sharp: high performers aren't just buying better chips; they are redesigning their entire operating models. They are finding that the most expensive part of AI isn't compute; it’s data debt and organizational inertia. If you’re still treating AI as a shiny plug-in for your existing inefficiencies, you aren't innovating; you're just automating them.
🧠 Explosive Adoption: GenAI usage has nearly doubled in a year, moving from niche pilot projects to core business functions.
⚡ Financial Impact: High-performing companies attribute over 10% of their top-line growth to AI-driven initiatives.
🎓 The Talent Pivot: Companies are moving away from hiring generalists toward specialists who can bridge the gap between LLMs and legacy data.
🔍 Scalability Crisis: The primary hurdle is no longer the technology itself, but the "data readiness" and governance required to move beyond the sandbox.
https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai
#StateOfAI #DigitalTransformation #Leadership #GenAI
Báo cáo "State of AI" phân tích 100 nghìn tỷ token cho thấy: AI chuyển từ tạo văn bản sang mô hình suy luận đa bước. Đặc biệt, mô hình mã nguồn mở đang bùng nổ, chiếm 1/3 tổng mức sử dụng và cạnh tranh mạnh mẽ. Các mô hình cỡ trung (15-70B) được ưa chuộng nhất. Ứng dụng phổ biến vẫn là lập trình, QA, năng suất.
#AI #OpenSourceAI #LLM #StateOfAI #XuHuongAI #MoiTruongMo #CongNgheAI
https://www.reddit.com/r/LocalLLaMA/comments/1penmet/key_insights_from_the_state_of_ai_report_what/
#StateOfAi report is out... and its going to take me a while to digest all 300 slides in the deck...
But here is one slide that is a counterpoint to the #Ai Antagonists and doomers who cry about the insance power requirements of AI engines...
My point: It is entirely unreasonable to expect the AI #Datacentres to maintain the same scalar demand for power and water growth, given every single other industry optimising as it progresses.
One such optimisation of #LLM is model distillation, the other COCONUT. Basically recycling tokens at inference time... thus greatly reducing compute burn...
An interesting downside is that we can not monitor what the AI is 'thinking' with COCONUT algo.
2025 State of AI Code Quality
https://www.qodo.ai/reports/state-of-ai-code-quality/
#HackerNews #StateOfAI #CodeQuality #AI2025 #TechTrends #QodoReport #AIInsights
State of AI Report 2023 - If there is ONE study you HAVE to read - competence to the point!