From Agent-1 to Superintelligence: Decoding the AI 2027 Scenario and Its Profound Implications Check out my article: linkedin.com/pulse/from-agen… Via @ingliguori #AI2027 #FutureTech
→ View original post on X — @ingliguori, 2026-04-02 12:17 UTC
RESEARCH
-
Slopsquatting: AI Hallucinations Enable New Supply Chain Attack
By
–
URGENT PSA – New supply chain attack vector that I found WILD > AI LLMs hallucinate package names roughly 18-21% of the time.
— Basel Ismail (@BaselIsmail) 2 avril 2026
Hackers have started pre-registering those hallucinated names on PyPI and npm with malicious payloads; they call it "slopsquatting"
You can only imagine… pic.twitter.com/wyPwHE9NT5URGENT PSA – New supply chain attack vector that I found WILD > AI LLMs hallucinate package names roughly 18-21% of the time. Hackers have started pre-registering those hallucinated names on PyPI and npm with malicious payloads; they call it "slopsquatting" You can only imagine what's next Community note: The 'slopsquatting' attack vector was documented as early as April 2025 and not newly discovered. The cited 18-21% package hallucination rate applies to open-source LLMs; commercial models average 5.2% according to the referenced study using pre-2025 models. socket.dev/blog/slopsquat… arxiv.org/pdf/2406.10279
-
Researcher Attribution Matters in Open Source AI Community
By
–
Proper researcher attribution is so important. The open source AI community runs on people getting credit for their work. Super cool to see this!
-
Sakana releases interesting updates with Marlin beta launch
By
–
Sakana keeps shipping interesting stuff. Will definitely check out Marlin beta!
-
RT-DETR: Real-Time Detection Transformers Revolution in Computer Vision
By
–
⚡RT-DETR: Real-Time Detection Transformers
— Satya Mallick (@LearnOpenCV) 2 avril 2026
Transformer detectors like DETR and DINO were powerful but too slow for real-time use. In 2023, RT-DETR changed the game. 🚀
By combining CNN backbones with efficient transformer attention and IOU-aware query selection, RT-DETR… pic.twitter.com/LntFCENHnJ⚡RT-DETR: Real-Time Detection Transformers Transformer detectors like DETR and DINO were powerful but too slow for real-time use. In 2023, RT-DETR changed the game. 🚀 By combining CNN backbones with efficient transformer attention and IOU-aware query selection, RT-DETR achieved strong accuracy while running at real-time speeds. ⚡ It proved that transformers could finally match CNNs not just in accuracy, but in speed—pushing object detection into a new era. #RTDETR #Transformers #ComputerVision #DeepLearning #AI #ObjectDetection #MachineLearning #AIResearch #DataScience 🤖
→ View original post on X — @learnopencv, 2026-04-02 11:15 UTC
-
AI Task Duration: Understanding Machine Learning Execution Times
By
–
#AI Task Duration
by @Khulood_Almani #ArtificialIntelligence #MachineLearning #ML -
PrismML’s 1-bit Bonsai Models Achieve 10x Intelligence Density Improvement
By
–
Pretty insane: PrismML has introduced its 1-bit Bonsai models, compressing an 8.2B-parameter LLM into just 1.15 GB while maintaining performance comparable to leading 8B models. By focusing on “intelligence density,” the company claims roughly a 10x improvement in capability per
-
From Fear to Familiarity: AI Adoption Psychology
By
–
Interesting how familiarity breeds confidence rather than fear here. Early 2023 was definitely the most "what is happening" moment for me too haha
-
Curating 1,900 Hand-Picked AI Newsmakers from 50,000 Followers
By
–
How can I pick one? I have 50,000 all in lists: https://
x.com/scobleizer/lis
ts
… My favorite list is AI Newsmakers, but that has 1,900 on it, all hand picked. -
Context Engineering: Organizing Unstructured Data for AI Reasoning
By
–
What is “Context Engineering”? It’s the discipline of stitching your unstructured mess—logs, chats, docs, images—into something an AI can actually reason over. → Vector DBs + hybrid search + embeddings to retrieve by meaning, not keywords. → Decisions anchored in your data,