Statistics reveal that 70% of businesses have not yet deployed agentic AI solutions to address real-world problems. Despite the surge in frameworks and tools for developing LLM-based agents, only a small fraction of these solutions are in production beyond the proof-of-concept stage. The terminology surrounding agentic AI remains inconsistent, though definitions from the Anthropic blog “Building Effective Agents” are gaining popularity. While agentic AI shows promise in open-ended web searching tasks, such as deep research where user validation is crucial, its application in autonomously navigating internal business systems remains largely untested and unreliable. The disconnect between the hype and practical implementation of agentic AI is significant, raising questions about its real-world utility and the challenges of debugging such systems.
Source: www.reddit.com

Related Videos
Related X Posts
metasal | acc/sol
@metasal_
·
2h
Some Agents from Szn 1 have been decommissioned it seems. No slop on the tl for months. Maybe the AWS bill arrived?
Pratyus Patnaik
@pratyus
·
Apr 8
Beneath all the Agentic AI excitement, one issue keeps coming up in conversations with IT leaders: Access management.For years, software architecture in enterprises has assumed humans in the loop. IGA, PAM, approval chains, ticketing workflows—all are designed around a person
Pratik Kadam
@PratikKadam_
·
Mar 7
My AI Agent Tech Stack• No-code: n8n
• Scraping: Firecrawl
• Cloud Provider: GCP
• Database: Supabase
• Deployment: Docker
• Vector Store: Pinecone
• Model Routing: Openrouter
• Agent Observability : Agentops
• Agent Orchestration: Langgraph
• Foundational Model :
Poonam Soni
@CodeByPoonam
·
Apr 3
OpenAI just got DeepSeek’d again.A tiny startup based out of London just dropped Agent Swarms and it’s INSANEHere’s EVERYTHING you need to know:
The New Stack
@thenewstack
·
19h
Agentic AI: The Missing Piece in Platform Engineering
@gitlab
Stanford NLP Group
@stanfordnlp
·
Feb 12
The final admission that the 2023 strategy of OpenAI, Anthropic, etc. (“simply scaling up model size, data, compute, and dollars spent will get us to AGI/ASI”) is no longer working!