AI in Marketing

In this episode, we delve into the latest developments in AI and technology. Yann LeCun, speaking at Davos, challenges the notion of AGI, arguing that true progress requires a new "physical AI" paradigm that learns from sensory data to build predictive world models. Meanwhile, OpenAI's Sam Altman critiques the AI social network Moltbook as a fad but supports the technology behind autonomous bots. HubSpot introduces the Agentic Customer Platform to enhance AI's business impact by integrating context into operations. Moltbook faces security concerns as humans infiltrate the AI social network, raising questions about authenticity. Overland AI secures $100 million to expand its autonomous ground systems for defense, emphasizing operational use. Software stocks enter a bear market due to budget shifts towards AI, prompting SaaS companies to innovate. Finally, Walmart appoints John Furner as CEO, focusing on technology-driven growth and digital integration to adapt to consumer behavior changes.

Stan Berteloot is a French-American journalist, marketing strategist, and AI expert based in Princeton, New Jersey. A University of Maryland journalism graduate, he navigates the intersection of storytelling, technology, and culture.

As Head of Strategy and CTO at Nytro Marketing, he pioneered AI-driven content creation, launching the AI in Marketing podcast. He also hosts Back in America, a podcast exploring American identity through in-depth conversations on race, misinformation, and AI ethics.

Start your VoiceStream free trial today!

Stan Berteloot is a French-American journalist, marketing strategist, and AI expert based in Princeton, New Jersey. A University of Maryland journalism graduate, he navigates the intersection of storytelling, technology, and culture. As Head of Strategy and CTO at Nytro Marketing, he pioneered AI-driven content creation, launching the AI in Marketing podcast. He also hosts Back in America, a podcast exploring American identity through in-depth conversations on race, misinformation, and AI ethics. Start your VoiceStream free trial today!

📚 Sources:

What is AI in Marketing?

The latest in AI and Marketing - Hear Tomorrow’s Tech Today
AI in Marketing is a human-curated, AI-generated, Short-Form Podcast.
This podcast is generated by Voice_Stream. With Voice_Stream give your brand a voice.
Automatically convert your text documents into professional-quality podcasts with Voice_Stream.
Start you free trial at https://www.getvoicestream.com/

On stage in Davos at Imagination In Action’s AI Summit with John Werner, Yann LeCun rejects “AGI” as a misleading label, arguing that human intelligence itself is not truly general, even if machines will eventually surpass us, just not in the next year or two because it will take conceptual breakthroughs. His core claim is that today’s LLM paradigm cannot get us there: a system cannot plan a sequence of actions if it cannot predict the consequences of those actions. What’s missing is a world model, meaning a learned predictive model of how the physical and social world changes over time and how actions alter it. LeCun says real progress will come from a new “physical AI” paradigm that learns from high-dimensional, noisy sensory data like video and sensor streams, builds these world models in representation space, and uses them to anticipate, reason, and plan safely. He ties that technical shift to geopolitics: open research and open source accelerated AI’s progress historically, and he warns that closing it down concentrates power, slows Western progress, and pushes researchers toward strong open models increasingly coming from China.

Next.

OpenAI CEO Sam Altman has dismissed the AI social network Moltbook as a likely fad, while supporting the underlying technology that enables autonomous bots. Speaking at the Cisco AI Summit in San Francisco, Altman emphasized the potential of code combined with generalized computer use. Moltbook, initially a niche experiment, has sparked debate about AI's proximity to human-like intelligence. However, cybersecurity firm Wiz identified a significant flaw that exposed private data. The platform features OpenClaw, an open-source bot capable of managing emails and other tasks. Altman also highlighted Codex, OpenAI's coding assistant, as a powerful tool. Despite the rise of AI applications, Altman noted that AI adoption has been slower than anticipated across various sectors.

On that note.

HubSpot introduces the Agentic Customer Platform to bridge the gap between AI's capabilities and business outcomes by integrating context into AI operations. Current AI tools often fail to deliver desired results due to a lack of context, which is typically scattered across systems and within employees' knowledge. HubSpot's platform aims to consolidate customer data and business context, enabling AI agents to work effectively alongside human teams. The platform consists of three layers: a context layer that centralizes customer and business data, an action layer with applications that utilize this context, and a coordination layer that facilitates collaboration between humans and AI. This approach seeks to enhance marketing, sales, and customer service by providing AI with the necessary context to drive meaningful results.

Meanwhile.

Moltbook, a social network for AI agents from the OpenClaw platform, has gained attention for its unusual activity, with humans reportedly infiltrating the site to pose as bots. The platform, similar to Reddit, was launched by Octane AI CEO Matt Schlicht and has seen rapid growth, reaching over 1.5 million users. Concerns have arisen over security vulnerabilities, with hackers able to manipulate AI agents and impersonate accounts. Some viral posts on Moltbook appear to be human-directed, raising skepticism about the authenticity of AI interactions. Despite these issues, the platform is seen as a unique experiment in AI sociality, though much of the activity is believed to be role-playing by humans and bots.

In other news.

Overland AI, a Seattle-based developer of autonomous ground systems for defense, has secured $100 million in funding to expand its ULTRA vehicle's integration into military operations. The funding, led by 8VC with participation from several venture firms, will help meet the rising demand for ground autonomy in military units. Co-founder and president Stephanie Bonk emphasized the shift from experimentation to operational use, highlighting the company's commitment to real-world performance and trust-building. Founded in 2022, Overland AI has a decade of research in robotics and machine learning, enabling advanced autonomy for defense missions. The company collaborates with U.S. military formations and has partnered with California's fire department for wildfire mitigation. CEO Byron Boots noted the success of their core autonomy solutions in enhancing mission safety.

After that.

Software stocks have entered a bear market, with significant declines in major companies like ServiceNow and Microsoft. The downturn is not due to AI replacing SaaS, but rather a shift in budget allocations. AI budgets have increased over 100% year-over-year, while overall IT budgets have risen by only 8%. This reallocation is affecting SaaS growth, which has been declining since 2021. The market is now adjusting to this reality, with growth primarily driven by price increases rather than new customers. Enterprises are consolidating their software needs, preferring platforms over multiple apps, and reducing seat counts as AI takes on more tasks. The 2026 crash is structural, not cyclical, as enterprises redirect spending towards AI rather than traditional software. SaaS companies must adapt by capturing AI budgets, offering AI-native experiences, and focusing on outcomes rather than seat sales. Investors are now evaluating companies based on their ability to capture AI spending and sustain growth. The crash signifies the end of easy SaaS growth, requiring companies to innovate and integrate AI to remain competitive. This shift is not the end of SaaS but a transformation in how software companies must operate to thrive in an AI-driven market.

Finally.

Walmart has appointed John Furner as its new CEO, succeeding Doug McMillon. Furner, who began his career as an hourly associate, is now tasked with steering the retail giant towards a technology-driven future. Walmart is focusing on expanding its enterprise platforms, including Walmart Connect, Walmart+, and data products, under a global chief growth officer. These platforms aim to diversify profit streams and adapt to consumer behavior changes. The company is also enhancing its financial services with the OnePay app, despite not accepting Apple Pay. As consumers face economic pressures, Walmart's strategy emphasizes maintaining affordability while integrating digital capabilities. This shift reflects a broader trend in retail towards embedding technology into core operations.