OpenAI Slows Down as Anthropic, Broadcom, Amazon, and India Face the Next AI Shift
AI is getting more powerful, but the pressure around it is getting harder to ignore.
This week, OpenAI slowed frontier-model development because of cybersecurity concerns, Anthropic reportedly hit a $65B revenue run rate, Broadcom looked for more than $60B in AI chip financing, and OpenAI launched ChatGPT for Teens with stronger safeguards.
In this episode of AI Edge by Mahesh Devalla EP15, we look at how the AI race is becoming more controlled, more expensive, and more deeply connected to business execution.
Who can build powerful AI, protect users, control infrastructure costs, and still turn AI into real outcomes?
Highlights from this episode
OpenAI slows frontier-model development as cybersecurity concerns rise
Anthropic hits a $65B revenue run rate as Claude’s commercial growth accelerates
Broadcom seeks more than $60B for custom AI chip infrastructure
ChatGPT for Teens launches with stronger safeguards and parental controls
Amazon reportedly destroys rare books for AI training data
India’s $315B IT-services model gets pushed away from headcount billing toward outcome-based work
Relay shuts down as the team joins Google Chrome
LinkedIn’s AI-slop button crosses 1M uses
The AI race is now about safety, revenue, chips, training data, services, browser strategy, and content quality
Quick takeaway
AI is entering a more serious phase.
The headlines are not only about faster models anymore. They are about cybersecurity limits, infrastructure financing, teen safety, data sourcing, enterprise business models, automation talent, and users pushing back against low-quality AI content.
The next phase will reward companies that can scale AI while keeping trust, cost, quality, and execution under control.
Watch the full video below.
For more AI news, business insights, and leadership analysis, follow Mahesh Devalla.