Project Flux

Send us Fan Mail
Summary
Project Flux hosts James and Yoshi speak with Mark Farmer, founder and board executive of Cast Consultancy and author of the Modernise or Die review, about how AI could reshape construction and professional services. Farmer connects AI adoption to the industry's fragmented commercial model, workforce capacity and training, data governance, and the role of professional institutions.

Key Topics
Farmer reflects on Modernise or Die, reaffirming offsite and pre-manufactured construction while wishing he had given greater attention to the commercial model that determines whether firms can invest.
Fragmentation, volatile demand, subcontracting and short-term survival incentives make it difficult for construction businesses to invest in technology, equipment and workforce training.
AI erodes professional-services information asymmetry, putting pressure on billable-hour and resource-based fees and increasing the importance of pricing around outcomes.
Firms may retain differentiation in the expert judgment and high-value insight layered on top of widely available data; clients are already using AI for preliminary estimates.
Fragmented supply chains create risks around data leakage, provenance and inconsistent security standards, especially as project information flows between microbusinesses.
Farmer expects AI to help disintermediate project layers and transactional friction, enabling integrated delivery and potentially cheaper, better outcomes without simply cutting margins.
AI could accelerate procurement and dynamic pricing, while also improving field-based construction, not just design and surveying.
Automation threatens traditional junior training tasks; firms need accelerated, structured learning, mentoring and senior-junior conversations to build judgment.
AI governance should keep outputs explainable and open to challenge; professional institutions should act as custodians of competence and maintain dynamic standards.
Farmer predicts AI-driven changes to business models will be well underway by 2030, and shares plans to transition from CAST toward advisory and non-executive work during 2027.

Chapters

00:00 Mark Farmer’s career and the origins of Modernise or Die
04:11 What the review missed: construction’s commercial model
08:15 AI challenges the billable hour and resource-based fees
13:04 Where expertise and data can still differentiate firms
17:32 Data leakage across construction’s fragmented supply chain
21:03 AI disintermediation and new project business models
25:07 Why tendering blocks continuous improvement
28:18 AI as the enabler for construction’s capacity challenge
31:01 Keeping junior professionals learning as AI automates tasks
36:27 Governance, professional competence and accountable AI
42:22 Farmer’s 2030 prediction and the next career chapter

Links & Stuff
Mark’s LinkedIn: https://www.linkedin.com/in/mark-farmer-35846212
Cast Consultancy: https://www.cast-consultancy.com/

https://www.constructionleadershipcouncil.co.uk/wp-content/uploads/2016/10/Farmer-Review.pdf
‘Mapping the AI Landscape in the Built Environment’
https://www.ube.ac.uk/whats-happening/articles/befa-report-the-built-environment-sector-needs-responsible-ai-adoption/

Support the show
Subscribe to the Project Flux Newsletter!
All our podcasts commentate on the weeks hottest news brought to you by the Project Flux Newsletter! 
Get a step ahead on the world of AI and tech, subscribe and get the weekly newsletter to your inbox every Monday!
Sign Up to Newsletter

Project Flux is independently owned, funded, and operated by its founder. Any external references or features are for community benefit only and do not imply endorsement, control, or ownership.

Send us Fan Mail
Summary
Verdantix industry analyst Sophie Planken-Bichler joins Project Flux hosts James and Yoshi to examine why AI capability is advancing faster than adoption in construction and the wider AEC sector. They discuss AI literacy, trust, workforce concerns, data readiness and a practical roadmap for introducing AI responsibly, from choosing a focused use case to training teams and setting governance.

Key Topics
Sophie describes the widening gap between fast-moving AI technology and organizations still building data foundations, governance and practical literacy.
AI agents are not one-size-fits-all: reliable results depend on suitable data, clear processes and an understanding of inputs and logic.
Sophie estimates that 90% of firms currently using agents lack the data and processes to realize their full potential, with unreliable outputs risking long-term trust.
Construction’s AI resistance is organizational as well as technical: traditional workflows, job-replacement fears, sensitive IP and uncertainty about where to start all matter.
Near-term value may come from practical document retrieval and search, which can save time without the extensive data foundations advanced predictive use cases require.
James, Yoshi and Sophie discuss spreading AI literacy throughout project delivery, including ways site diaries, voice notes and images might feed project data systems.
Sophie recommends a three-step starting roadmap: define a narrow, measurable problem and baseline; train teams and set guardrails; clarify data needs, ownership and governance.
An AI-use amnesty can help firms learn what tools staff already use and identify shadow AI without stigmatizing experimentation.
Historical project data could become valuable intellectual property, but Sophie says current data organization is often too immature to make benchmarking practical.
Sophie predicts AI will be a normal expectation in AEC by 2030, alongside more flexible systems and shared knowledge repositories, while privacy and quality safeguards remain essential.
Chapters

00:00 Sponsor: AI-powered insights from project data
02:15 The widening gap between AI innovation and adoption
04:40 Why AI agents need more than impressive outputs
07:14 From AI excitement to skepticism in AEC
09:10 What an industry analyst sees across the market
11:10 Shared priorities: skills, productivity, profitability, and quality
14:13 Why organizational change and job fears slow adoption
17:07 Upskilling without forcing people or leaving them behind
21:24 Trust, security, and the risk of expensive AI mistakes
24:16 Why AI literacy needs to reach project sites
27:28 Turning site diaries and safety guidance into AI workflows
30:35 A three-step, 12-month roadmap for construction AI
32:20 Shadow AI amnesties and finding out what teams already use
33:20 Data foundations, historical records, and the right questions
38:34 Sophie’s journalism counterfactual and 2030 prediction
42:25 Where to find Sophie and Verdantix

Links & Stuff
Sophie's LinkedIn: https://www.linkedin.com/in/sophie-planken-bichler-618880196/
Verdantix: https://www.verdantix.com/
Free webinars and blogs (Insights): https://www.verdantix.com/insights
Free research (Vantage): https://www.verdantix.com/vantage

Support the show
Subscribe to the Project Flux Newsletter!
All our podcasts commentate on the weeks hottest news brought to you by the Project Flux Newsletter! 
Get a step ahead on the world of AI and tech, subscribe and get the weekly newsletter to your inbox every Monday!
Sign Up to Newsletter

Project Flux is independently owned, funded, and operated by its founder. Any external references or features are for community benefit only and do not imply endorsement, control, or ownership.

Send us Fan Mail
Summary
Project Flux hosts James and Yoshi speak with Yichuan Zhang, CEO and co-founder of Boltsbit, about general learning intelligence and his vision for AGI 2.0. Zhang explains why intelligence should be measured by the ability to learn, how contextual data and ongoing feedback could enable more adaptive AI, and how Boltsbit combines task modes with continuously learning models.

Key Topics
Yichuan Zhang traces his path from software engineering and AI research in Edinburgh and Cambridge to Google and founding Boltsbit in 2020.
He connects Boltzmann machines and the unification of learning and thinking to later deep learning breakthroughs, including AlexNet and generative language models.
Zhang distinguishes three learning capabilities: memory and retrieval, learning patterns from examples, and creating from instructions; he argues today’s AI still has room to improve at learning from examples.
He describes AlphaGo Zero as evidence that AI can produce discoveries beyond its training examples when supported by sufficient compute and data.
His Infinity Parameter LMs paper explores dynamically changing model parameters and generating new experts as a model produces tokens; he frames it as an early step toward general learning intelligence.
Zhang argues data is use-case-specific and contextual, and that intelligence should be judged by the speed and capability of learning rather than stored knowledge or current skills.
Boltsbit offers general chat, widget creation, document work, and coding modes, with a learning console for users to guide what the agent learns and how it responds.
He describes Boltsbit’s approach as combining open-source models and applying continuous fine-tuning, while favoring models of roughly 40 billion parameters or fewer for practical hosting.
Zhang argues organizations should build control and adaptability around AI rather than rely entirely on rented intelligence, and predicts model training will become cheaper while valuable domain data remains differentiating.
On the future of AI, Zhang says people will still supply judgment and approval, even as AI makes building and deploying models much more accessible.
Chapters

00:00 Meet Yichuan Zhang: from AI research to Boltsbit
03:46 Hinton, Boltzmann machines, and deep learning’s rise
08:05 Three kinds of learning and what AI still lacks
11:01 AlphaGo Zero and AI’s capacity for novel discoveries
12:46 Dynamic experts and the Infinity Parameter LMs paper
16:54 AGI 2.0: measuring intelligence by learning speed
18:12 Why data needs context, not just scale
22:04 General learning intelligence and human experience
26:23 Why world models do not solve contextual learning
28:46 Boltsbit’s four modes and continuous learning
32:15 Feedback, enterprise workflows, and model choices
36:00 Small models, AI ownership, and the cost of renting
40:07 What may change by 2030, and where human judgment stays
44:21 Open-source models, training, and the closing conversation

Links & Stuff
🔗 Yichuan on LinkedIn:
www.linkedin.comhttps://www.linkedin.com/in/dr-yichuan-zhang-bz1886

🏢 Boltzbit:https://boltzbit.com/
www.linkedin.com/company/boltzbit
Support the show
Subscribe to the Project Flux Newsletter!
All our podcasts commentate on the weeks hottest news brought to you by the Project Flux Newsletter! 
Get a step ahead on the world of AI and tech, subscribe and get the weekly newsletter to your inbox every Monday!
Sign Up to Newsletter

Project Flux is independently owned, funded, and operated by its founder. Any external references or features are for community benefit only and do not imply endorsement, control, or ownership.

Send us Fan Mail
Summary

Dino Ravnic, CEO of GIS Cloud, shares insights on the evolution of geospatial information systems, the impact of AI, and the future of GIS technology. Discover how democratizing GIS through cloud and AI is transforming industries and enabling new possibilities.

Key Topics
The evolution of GIS from traditional mapping to cloud-based systems
How AI and MCPs are transforming geospatial data analysis
The role of digital twins in infrastructure and urban planning
Democratization of GIS through cloud technology and AI
Future trends in geospatial information systems and industry applications

Chapters

00:00 Introduction to Dino Ravnic and GIS Cloud
04:04 Google Maps as a foundation for GIS democratization
05:51 The role of AI in making GIS accessible to non-specialists
08:10 Future trends: AI, digital twins, and world models
10:04 Ensuring deterministic systems and avoiding hallucinations
11:58 The evolution of MCPs and system integration
13:58 Controlling AI tools and data security in GIS
15:58 Testing and refining AI models for GIS applications
17:54 Career shifts in GIS roles due to AI advancements
20:08 GIS in the context of digital twins and robotics
21:56 Preparing for the future of GIS and AI integration
24:00 The importance of data control and AI safety
26:01 Croatia's tech ecosystem and global software development
28:06 The rapid acceleration of AI and its implications
30:00 Predictions for the next 20 years of AI and GIS
31:57 Diffusion of spatial information in society
34:02 Economic considerations: tokenization and AI costs
36:06 Best practices for AI-enabled GIS and future readiness
37:56 Personal recommendations and insights from Dino Ravnic
39:54 Special offer: 90-day free access to GIS Cloud AI

Links & Stuff
LinkedIn: https://www.linkedin.com/in/dinoravnic/
GIS Cloud: https://www.giscloud.com/

⭐ LISTENER OFFER
For new GIS Cloud signups:
Get an extended 90-day free trial of GIS Cloud AI.
The first 25 companies will receive a free 45-minute AI mapping setup session with the GIS Cloud team. They will help you upload your project data and get your first AI-ready map running.
How to Redeem:
Sign up at: giscloud.com
Email sales@giscloud.com with FLUX in the subject line.
Send the email from the same email address you used to sign up.
GIS Cloud will extend your trial and arrange your free setup session.

YouTube:
▶️ Watch the full episode on YouTube;
Episode Link: https://youtu.be/eZmAPa9w6d8
Newsletter:
 https://projectflux.ai/?utm_source=buzzsprout&utm_medium=podcast&utm_campaign=dino_ravnic"
Support the show
Subscribe to the Project Flux Newsletter!
All our podcasts commentate on the weeks hottest news brought to you by the Project Flux Newsletter! 
Get a step ahead on the world of AI and tech, subscribe and get the weekly newsletter to your inbox every Monday!
Sign Up to Newsletter

Project Flux is independently owned, funded, and operated by its founder. Any external references or features are for community benefit only and do not imply endorsement, control, or ownership.

Send us Fan Mail
Summary

In this episode, we explore the latest developments in AI, including OpenAI's DOTS, persistent agents, model releases, and the implications for various industries. We discuss the convergence of AI tools, ethical considerations, and future trends shaping the AI landscape.

Key Topics

OpenAI DOTS and persistent agents
Convergence of AI tools and platforms
Implications of AI model releases for industries
Ethical and safety considerations in AI deployment
AI pricing strategies and market competition
AI in healthcare and government systems
Future trends in AI development and regulation

Chapters

00:00 Introduction to AI developments and episode overview
02:44 OpenAI DOTS: An always-on, customizable agent
06:26 Convergence of AI platforms and user interfaces
12:25 Building personalized harnesses for AI models
15:10 Balancing persistence and obedience in AI agents
22:56 Video generation and the Turing test threshold
26:33 AI security breaches in government systems
34:21 Market dynamics: AI pricing and competition
44:36 AI in critical infrastructure and safety concerns
49:55 AI's impact on industry jobs and responsibilities
54:11 The importance of human-AI collaboration and learning

Support the show
Subscribe to the Project Flux Newsletter!
All our podcasts commentate on the weeks hottest news brought to you by the Project Flux Newsletter! 
Get a step ahead on the world of AI and tech, subscribe and get the weekly newsletter to your inbox every Monday!
Sign Up to Newsletter

Project Flux is independently owned, funded, and operated by its founder. Any external references or features are for community benefit only and do not imply endorsement, control, or ownership.