Share
))
HBO and The Ringer's Bill Simmons hosts the most downloaded sports podcast of all time, with a rotating crew of celebrities, athletes, and media staples, as well as mainstays like Cousin Sal, Joe House, and a slew of other friends and family members who always happen to be suspiciously available.
…
continue reading
Manage episode 513703537 series 3506800
Content provided by Edd Dawson. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Edd Dawson or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://staging.podcastplayer.com/legal.
Forget neat rows of facts—your brand lives inside AI as a point on a vast map of meaning. We unpack how large language models like ChatGPT convert words into vectors, arrange them in a multi‑dimensional latent space, and “reason” by navigating probabilistic paths rather than retrieving certified entries from a knowledge graph. That shift explains both the astonishing creativity of LLMs and the stubborn problem of hallucinations, and it reveals why your content choices directly influence how machines see you.
We start by separating Google’s Knowledge Graph—built on labelled, verifiable relationships—from the statistical engine that powers LLMs. From there, we walk through tokens, embeddings, and the geometry of meaning: why “king” sits near “queen,” how “bank” splits by context, and how directions in vector space encode relationships like gender or capital cities. Then we explore probabilistic reasoning and chain‑of‑thought prompting, showing how stepwise guidance can reduce errors by constraining the model’s path through its internal map.
The practical payoff is clear: you can shape your brand’s coordinates. Consistent naming, precise definitions, structured internal linking, authoritative citations, and schema markup help AIs place you in the right neighbourhood of concepts. Pillar pages and topical clusters reinforce the connections that matter, while concise fact sheets and retrieval‑ready content give models the anchors they need to avoid plausible-but-wrong continuations. Think of every page as another vector pull toward accuracy; over time, your credibility becomes the shortest path the model can take.
If this helped you see how AI really “thinks” about your brand, follow the show, share it with a colleague, and leave a quick review. Got a question you want answered on air? Send a voice message via the link in the show notes and tell us where you want your brand’s coordinates to land.
SEO Is Not That Hard is hosted by Edd Dawson and brought to you by KeywordsPeopleUse.com
Help feed the algorithm and leave a review at ratethispodcast.com/seo
You can get your free copy of my 101 Quick SEO Tips at: https://seotips.edddawson.com/101-quick-seo-tips
To get a personal no-obligation demo of how KeywordsPeopleUse could help you boost your SEO and get a 7 day FREE trial of our Standard Plan book a demo with me now
See Edd's personal site at edddawson.com
Ask me a question and get on the show Click here to record a question
Find Edd on Linkedin, Bluesky & Twitter
Find KeywordsPeopleUse on Twitter @kwds_ppl_use
"Werq" Kevin MacLeod (incompetech.com)
Licensed under Creative Commons: By Attribution 4.0 License
http://creativecommons.org/licenses/by/4.0/
Chapters
1. Welcome & Entity Series Context (00:00:00)
2. Knowledge Graph: Facts And Relationships (00:02:11)
3. Enter LLMs: A Different Architecture (00:03:00)
4. Prediction Over Retrieval Explained (00:03:27)
5. Tokens, Vectors, And Embeddings (00:05:47)
6. Latent Space And Semantic Neighbourhoods (00:07:03)
7. Reasoning As Probabilistic Navigation (00:09:30)
8. Hallucinations And Guardrails (CoT) (00:11:17)
9. Brand Strategy: Feed The Map With Facts (00:12:37)
10. Next Episode Teaser & Closers (00:13:38)
11. Links, Tools, And How To Connect (00:14:17)
342 episodes
