Lutz Finger

Executives: Stop asking how your brand ranks in AI. The real question is: Why should AI trust your content and brand over others? Are you the definitive answer to your customer's questions? #AI #MarketingStrategy

Marketing is evolving with agent-to-agent communication (A2A). Agentic workflows are key to navigating the new GEO shift. Join our workshop: ecornell.cornell.edu/courses/artifi… #Marketing #AI #Automation #A2A #GEO

To help AI understand your brand and recommend it, clearly list your products, services, and customer pain points. Explicit language in trusted places is key for AI to understand your brand. @CornellMBA @trymeridian @alexmdees #GEO #BrandBuilding #AIStrategy

Your customers’ questions are a goldmine. Build content around them. Build community around them. Specific. Authentic. Question-focused. That is how brands win #AI #citations. @CornellMBA @trymeridian @alexmdees #BrandBuilding #Community

#GEO is not about pages. It is about #trusted answers. Reddit leads citations today. LinkedIn may be the smarter long-term bet: more trust, more identity, less bot sludge. @CornellMBA @trymeridian @alexmdees#SocialMedia #AI #SEO

SEO was about being found by people. GEO is about being understood by machines and trusted enough to be recommended. If AI cannot understand your brand, your customer may never see it. @CornellMBA @trymeridian @alexmdees #SEO #GEO #DigitalMarketing

OpenAI ads are not that simple. Google ads worked because everyone wanted the user to click out. ChatGPT wants to keep the user in. That is why the incentive model gets messy fast. @CornellMBA @trymeridian @alexmdees @CornellMBA @trymeridian @alexmdees #OpenAI #Ads #AI

How do LLMs learn. They find what is new and novel. A coffee tasting like peach instead of chocolate, grabs an LLMs attention. It's all about novel concepts for AI. @CornellMBA @trymeridian @alexmdees #AIMarketing #LLMs #ProductMarketing

Instead of voting, we submit reward functions that guide AI agents. Society's collective reward functions train the AI, ensuring it aligns with everyone's values. It's like replacing politicians with AI. Thx @CornellMBA #AI #Governance #FutureTech

Your next customer may never visit your website. Their AI will. The website is not dead. But the machine is now the first user. @CornellMBA @trymeridian @alexmdees #GEO #AI #WebDesign

High-intent traffic from sources like ChatGPT converts 20x more than general Google searches because users have already decided. Prioritize quality leads over quantity.@trymeridian @alexmdees #MarketingStrategy #SEO #GEO

#SEO got you found. #GEO gets you understood, trusted, and recommended by #AI. Your homepage is no longer the front door. The model is. Thanks @alexmdees and @trymeridian for an amazing discussion @CornellMBA - ecornell.cornell.edu/keynotes/overv…

Role of #education in a world of #AI. We must teach how to question, verify, and work with AI without surrender logic and #observability. Good study from @Wharton linkedin.com/posts/lutzfing…

AI memory can be hacked. Prompt injection is not just a bad chat input — it can become long-term context. And once context becomes truth, GEO turns into memory warfare. @CornellMBA #AI #TechEthics

Love affair with GPT or Gemini? Guardrails can bend. Memory won’t. No RAG, no persistent memory, no romance. Context window closes. It forgets you. #AI #TechExplained

Guard rails and prompts? You will soon get your 'Bitter Lesson: short-term AI gains from human knowledge eventually create constraints - AI learns directly from patterns. @CornellMBA #AI #MachineLearning #BitterLesson

Autonomous weapons: AI deciding *who* to target vs. *how* to achieve a goal. The latter, agentic AI, could lead to disaster, like eliminating a city to find one person. Some AIs refuse such tasks; others build code guardrails. Which path is safer? #AIdangers #FutureTech #Ethics

What can agents do? What tools are allowed? Where to stop or escalate? This choreography of small, structured steps defines agents and what we're learning now. #AI #Agent

AI’s biggest flaw? It doesn’t know when to stop. Agents can optimize forever, but still lack the judgment of “good enough.” AGI is far off. @CornellMBA #AI #Agents #Innovation

Unleash Your Inner AI Agent. Stuck on 'how’? Join me for my no-coding hackathon / workshop @CornellMBA . You will build in no time. ecornell.cornell.edu/courses/artifi…