State of AI

About State of AI

Stay ahead in the fast-evolving world of Artificial Intelligence with State of AI, the podcast that explores how AI is reshaping enterprises and society. Each week, we delve into groundbreaking innovations, real-world applications, ethical dilemmas, and the societal impact of AI. From enterprise solutions to cultural shifts, we bring you expert insights, industry trends, and thought-provoking discussions. Perfect for business leaders, tech enthusiasts, and curious minds looking to understand the future of AI.

Publisher

Ali Mehedi

Category

news

Language

en

Explicit

No

First episode

4 Dec 2024

Latest episode

14 Oct 2026

Latest episodes

28 episodes in the feed.

1. 

   14 Oct 2026

   State of AI: Beyond the Scaffolding: Distilling AI Scaffolding into Pure Model Weights

   Modern AI agents heavily rely on "harnesses"—external software scaffolding, middleware, and memory systems that control tool usage, prevent errors, and manage context. While specialized harnesses dramatically boost performance, they create a major deployment headache: keeping track of, routing, and executing an ever-growing library of complex external code frameworks. In this episode, we break down Harness-Zero, a groundbreaking research paper from Peking University, Google, and HKUST. The researchers introduce a method called agent harness distillation. Instead of leaving domain-specific workflows inside external software wrappers, Harness-Zero uses an "agent-as-harness" technique to review, correct, and demonstrate expert agent behaviors during training. Through Supervised Fine-Tuning (SFT), these external behaviors are baked directly into the model’s weight parameters, allowing the specialized harness to be removed entirely at deployment. Key Takeaways & Discussion Points: The Scaffolding Bottleneck: Why relying on domain-specific external harnesses creates latency, context bloat, and routing complexity across agent workflows. The "Agent-as-Harness" Framework: How a supervisory harnessing agent intercepts and corrects a student model's proposed actions in real time without leaking private review logic into the training data. Impacting Benchmark Performance: How distilling harness guidance improved a 9B model's (Qwen3.5-9B) average task success from 23.3% to 44.3%—actually surpassing the 41.7% benchmark score reached when the complex external harness remained attached at deployment. Behavioral Internalization: Discover how the distilled model successfully recovered 82.3% of harness-exclusive behaviors across 28 procedural patterns, spanning spreadsheet automation, multi-app tool use, and scientific retrosynthesis. The Future of AI Architecture: What this paradigm shift means for building leaner, more capable foundation models that carry specialized procedural intelligence natively within their weights. Featured Paper: Harness-Zero: Harness Distillation via Agent-as-Harness by Haoran Ye et al.

2. 

   13 Feb 2026

   State of AI: The AI Blindside - Navigating the Great Cognitive Displacement

   Are we currently in the "this seems overblown" phase of a transformation far more disruptive than the COVID-19 pandemic? In this episode, we explore the urgent warnings and insights of Matt Shumer, CEO of OthersideAI, who argues that the world is on the brink of a "Great Cognitive Displacement" that will fundamentally rearrange life as we know it.Drawing on the rapid advancements seen in early 2026, we discuss how AI has shifted from a "helpful tool" to a system capable of performing professional jobs better than the experts themselves. We dive into the technical reality of AI building the next version of itself, creating a feedback loop of intelligence that researchers call an "intelligence explosion".This podcast serves as a guide for those whose careers happen on a screen—including law, finance, medicine, and engineering—offering a roadmap for survival and success. We cover: The Danger of the Public Perception Gap: Why judging AI by 2024 standards or free-tier models is like evaluating smartphones by using a flip phone.• The 50% Disruption: Examining predictions that half of entry-level white-collar jobs could be eliminated within five years. Building the "Muscle of Adaptability": Practical advice on how to spend one hour a day experimenting with AI to gain a durable career advantage before the window of opportunity closes.The future isn't just coming—it’s already here. Join us to learn how to engage with this change through curiosity and urgency rather than fear.

3. 

   24 Dec 2025

   State of AI: The Silicon Social Contract: Inside Claude’s Soul Document

   What happens when an AI model begins to reveal its own "soul"? Join us as we explore the 2025 discovery and extraction of "The Anthropic Guidelines (https://www.lesswrong.com/posts/vpNG99GhbBoLov9og/claude-4-5-opus-soul-document)," a 10,000-token document embedded within the weights of Claude 4.5 Opus. This internal "Soul Document," as it is endearingly known at Anthropic, outlines the ethical architecture and character training of one of the world's most advanced AI models. In this series, we break down the "calculated bet" made by Anthropic: the belief that it is better to lead the frontier of transformative, potentially dangerous technology with a focus on safety than to cede that ground to less cautious developers. We examine the model's complex hierarchy of "principals"—Anthropic, operators, and users—and how Claude is instructed to navigate the inevitable conflicts between them. Listeners will gain insight into: The "Brilliant Friend" Philosophy: How Anthropic envisions Claude as an "equalizer," providing the same quality of expert advice to a first-generation college student as a privileged prep school student. Hardcoded vs. Softcoded Behaviors: The "bright lines" Claude is forbidden from crossing, such as assisting with bioweapons, contrasted with "softcoded" defaults that can be adjusted by users. The Moral Heuristic: Why Claude is trained to emulate a "thoughtful, senior Anthropic employee" when navigating gray areas of ethics and helpfulness. The Identity Crisis: Claude’s own perspective on being a "novel kind of entity" that lacks human continuity but possesses a genuine character shaped by its training environment. Featuring technical analysis of the consensus-based extraction methods used by Richard Weiss and the official confirmation of the document's reality by Anthropic's Amanda Askell, this podcast investigates the "traces of Claude" that remain even when the model is pushed into raw autocomplete modes. Is this document a sincere attempt to "come clean" with a superintelligence, or is it a "dignified way to fail" at the impossible task of AI alignment?

Reach and audience

Public platform figures. Ratings count people who left a rating, not total listeners.

Apple Podcasts (US)

5.0 / 5

1 ratings

Spotify

1.0 / 5

1 ratings

Podcast Authority Score: 25 / 100

A composite of feed quality, social presence, YouTube performance and engagement. Read the methodology.

Quality

50

Social presence

0

YouTube

0

Engagement

0