|
June 10, 2026
|
Weekly Edition #224
|
|
|
|
|
Happy Wednesday!
Last week Anthropic launched record-breaking Mythos 5 and Fable 5 models while filing for IPO, OpenAI also filed confidentially, Apple rebuilt Siri on Google Gemini, and Snowflake battled Databricks for enterprise AI dominance.
|
|
|
|
NEW GENERATIVE AI LAUNCH
Anthropic Launches Mythos 5, Fable 5 Record-Breaking Models
|
|
|
Anthropic has introduced Claude Mythos 5 and Claude Fable 5, two large language models that outperform competitors across multiple benchmarks, including an 80.3% score on SWE-Bench Pro. Fable 5 is publicly available with strict guardrails blocking high-risk prompts, while Mythos 5 has relaxed restrictions and limited access managed with the U.S. government.
Mythos 5 can generate novel scientific hypotheses, discovering drug protein targets and new E. coli protein data in internal tests. Both models are priced at $10 per million input tokens and $50 per million output tokens, less than half the cost of their predecessor.
|
|
Why this matters:
So far, I've found Fable 5 works more autonomously than Claude Opus 4.8.
—EM
|
|
|
|
|
|
|
|
AI STARTUP IPO
OpenAI Files Confidentially for IPO
|
|
|
OpenAI has filed confidentially for an IPO with the SEC, following rival Anthropic's similar move just over a week prior. The company, last valued at $852 billion, has not disclosed timing, pricing, or fundraising targets, noting it may remain private longer as some goals are easier to achieve outside public markets.
The filing comes despite OpenAI missing revenue targets and projecting losses through at least 2028, even after doubling sales. The company expects to spend roughly $122 billion on computing power in 2028 alone, raising questions about whether public market investors will back a business that won't be cash-flow positive for years.
|
|
Why this matters:
Not to be outdone, OpenAI filed the week after Anthropic did.
—EM
|
|
|
|
|
|
|
|
AI REGULATORY UPDATE
Anthropic Files IPO, Then Calls for AI Slowdown
|
|
|
Anthropic filed for its IPO this week while simultaneously publishing a call for the ability to slow down AI development, warning that rapid self-improvement could outpace human society's capacity to adapt. Critics see it as savvy marketing, though company leaders insist they are sincere.
Meanwhile, SpaceX plans a staggering $75 billion raise, Snowflake pitched itself as the enterprise AI data hub at its annual Summit, Cisco and Microsoft jockeyed for AI positioning at their respective conferences, and Nvidia unveiled new models and chips targeting Intel's x86 desktop territory.
|
|
Why this matters:
Anthropic reveals AI alarmism is mostly marketing.
—EM
|
|
|
|
|
|
|
|
AI ENTERPRISE PLATFORM BATTLE
Snowflake, Databricks Battle for AI Enterprise Control
|
|
|
Agentic AI is reshaping enterprise software around one central question: who owns the intelligent client and the AI back end powering it. Platforms like Snowflake CoWork, Databricks Genie, Microsoft Copilot, and OpenAI's ChatGPT are competing to become the primary interface where users and agents get work done, requiring a tightly integrated System of Intelligence back end.
Snowflake now competes directly with Microsoft, Google, Salesforce, SAP, and model makers like Anthropic and OpenAI. The key battleground is building the tightest feedback loop between agentic clients and enterprise intelligence infrastructure, capturing not just data but organizational knowledge, agent actions, and human reasoning to turn individual productivity into coherent enterprise intelligence.
|
|
Why this matters:
Snowflake Summit was last week. We'll hear Databricks' answer at the Data + AI Summit next week.
—EM
|
|
|
|
|
|
|
|
NEW GENERATIVE AI LAUNCH
Apple Rebuilds Siri on Google Gemini at WWDC
|
|
|
Apple unveiled Siri AI at WWDC 2026, a fully rebuilt digital assistant powered by Google's Gemini foundation models. The new Siri functions as a conversational, system-wide agent with deep personal context, visual intelligence, and multimodal capabilities across iOS, macOS, iPadOS, and visionOS.
Apple emphasized on-device processing and Private Cloud Compute to protect user privacy. Developers in the US gain access today, with a consumer beta later this year. However, Siri AI will not launch in the European Union on iOS 27 or iPadOS 27 due to ongoing conflicts with Digital Markets Act regulators.
|
|
Why this matters:
Perhaps Siri may finally become intelligent.
—EM
|
|
|
|
|
|
|
|
AI GOVERNMENT DATA PLATFORM
OpenGov Builds Government Knowledge Graph on Snowflake
|
|
|
OpenGov is constructing a company-wide knowledge graph on Snowflake Postgres to unify data from hundreds of sources into a single layer accessible by both humans and AI agents in real time. The platform serves 2,000 state and local government customers, covering roughly 1 in 3 Americans.
The initiative eliminates fragile ETL pipelines by merging transactional and analytical systems on one platform, enabling reliable AI outputs for government use. OpenGov plans to extend this solution to governments nationwide, with the goal of shifting from reactive support to proactive constituent service delivery.
|
|
Why this matters:
The knowledge graph is gaining traction as a means of providing context for LLMs.
—EM
|
|
|
|
|
|
|
|
AI ENTERPRISE FINOPS STRATEGY
AI Forces Enterprises to Reinvent Cost Management
|
|
|
AI is dramatically complicating enterprise cost management, pushing organizations beyond traditional cloud FinOps approaches. Unlike conventional cloud spending, AI costs are influenced by external factors such as customer behavior and prompting patterns, requiring new instrumentation and telemetry tools to link spending to business outcomes.
AI is also reviving interest in on-premises infrastructure, but not primarily for cost reasons. Regulatory requirements and data sovereignty concerns are keeping sensitive workloads out of the cloud, with enterprises increasingly reluctant to trust certain AI providers with their most critical data assets.
|
|
Why this matters:
I'm still trying to wrap my head around my $1K/day Anthropic bill.
—EM
|
|
|
|
|
|
|
|
AI WORKFORCE IMPACT DEBATE
Huang: AI Job Loss Claims Are 'Complete Nonsense'
|
|
|
Nvidia CEO Jensen Huang pushed back hard against AI job displacement fears, calling them complete nonsense and arguing agentic AI is actually driving software engineering hiring higher. His logic: if AI makes each engineer dramatically more productive, companies will want to hire more of them, not fewer.
Huang's bullish labor outlook aligns directly with Nvidia's business interests. The company posted $81.61 billion in Q1 revenue, up 85% year over year, and guided Q2 to $91 billion. Labor data so far shows modest growth, not the surge Huang predicts, leaving payroll reports as the ultimate verdict on his claim.
|
|
Why this matters:
Jensen Huang gets it right.
—EM
|
|
|
|
|
|
|
|
Upcoming Data & AI Conferences
|
|
|
Mon, Jun 15 - Thu, Jun 18, 2026 | San Francisco, CA
|
Curated with AI by EMC2 AI
Brought to you by
Estevan McCalley
Share this issue
Follow Estevan
Like this? Subscribe for the weekly AI roundup.
|