▶ 1:31 · $SPCX · Elon Has a Real Plan for Grok Now

episode briefing

Elon Has a Real Plan for Grok Now

2026-07-10 · 8 company · 5 thematic

sentiment

6 bull1 bear1 neu

speakers

Co-host of Limitless Podcast (episode 210), a 4x/week show covering the frontier of AI and technology with a newsletter reaching hundreds of thousands. The podcast recently spun out as an independent entity.

ejaaz

shan

Cohost specializing in AI applications, image/video generation tools, and designer workflows; evaluates practical utility of new model releases.

quote

“They're kind of classifying this as Opus class on a relative basis compared to Claude Opus 4.8 And it seems to have actually accomplished that on just about all of the benchmarks. In fact, it even exceeds it slightly in some benchmarks for…”

—unknown

now playing · $SPCX

$SPCX···bullish· highunknown

SpaceX AI returns to frontier with Grok 4.5 after $60B Cursor acquisition

SpaceX AI leveraged Cursor's code routing data moat and 500K GPU infrastructure to launch Grok 4.5, an Opus 4.8-class model at 1.4T parameters that exceeds benchmarks in coding ef…

$CURSORbullish· highunknown

Cursor's $60B acquisition by SpaceX AI validates code routing data moat for model training

Cursor's platform data — routing prompts across models for cost-efficient code generation — provides a unique training signal for how models think, which SpaceX AI used to rapidly…

$OPENAIbullish· mediumjosh kale

OpenAI launches GPT 5.6 with breakthrough voice mode and taste, GPT 6 imminent

GPT 5.6 solves the 'taste' problem making it feel more human-like, while the new full-duplex voice model (GPT Live 1) enables natural interruption and parallel processing, positio…

$DEEPSEEKbullish· mediumejaaz

DeepSeek building custom ASIC chips for vertical integration, mirroring OpenAI's Jalapeno

DeepSeek and Jepu are developing purpose-built accelerator chips to optimize their model stack, reduce reliance on Nvidia/Huawei, and achieve vertical integration — a coordinated…

$NVDA···bearish· mediumjosh kale

Export controls backfire: China forced to build sovereign chip stack, creating siloed ecosystems

US GPU export controls accelerated Chinese labs' custom chip programs (DeepSeek, Jepu, Huawei), leading to a fully siloed hardware supply chain within a year; Chinese models may c…

$META···neutral· mediumejaaz

Meta's Muse Image/Video uses agentic reasoning and own data centers for cost-efficient niche play

Meta's new models employ agentic reasoning — web search, self-critique, regeneration — running on their own massive data centers for cheaper token serving, but remain far from LLM…

$BYTEDANCEbullish· mediumshan

ByteDance's Cream image model challenges GPT Image 2 with Photoshop-like editing at low cost

Cream enables granular, layer-based image editing (swap elements, change palettes, separate objects) at a fraction of competitor cost, targeting professional designers rather than…

$NET···bullish· mediumunknown

Cloudflare launches AI agent micropayments gateway as 52% of crawler traffic is now AI training

With AI crawlers generating 100-10,000x more requests than referred traffic, Cloudflare's new payments API enables automated micropayments for content access, creating an economy…

←