Freelance technology reporter. Covering AI and cybersecurity. Feel free to send news tips to [email protected]
Joined October 2020
- Tweets105
- Following73
- Followers1.1K
- Likes45
Mercor and 1Password built an AI agent with Codex to investigate engineering incidents. The end result was a 57% lower time to mitigate. Adarsh Hiremath, co-founder and co-CEO of Mercor and Nancy Wang, CTO of 1Password walk me through the project below.
forbes.com/sites/timkeary/20…
Top engineers at Tinder are spending $15K a month on tokens.
Earlier this month I got the chance to connect with Vinay Kuruvila, CTO at Tinder, who walked me through how the company is "using AI at every stage of the dating journey."
observer.com/2026/09/tinder-…
AI agents are now performing cyber attacks. Just last week, Hugging Face experienced a breach after an autonomous agent exploited two code-execution paths to gain access to internal data and credentials.
Following the breach, I reached out to Hugging Face for comment, and co-founder and CEO Clement Delangue @ClementDelangue, told me that the incident confirms what many had already suspected; attackers are already using AI agents.
The lesson from the incident? That "defense needs to become agentic too."
forbes.com/sites/timkeary/20…
Earlier this month, I got the chance to speak to Steven Bailey, CFO of Match Group, about how the dating giant is managing its AI transformation across apps like Tinder and Hinge -- and most importantly, how it plans to pay for all this.
observer.com/2026/07/match-g…
Tim Keary retweeted
Everyone is fighting over AI models. Bezos is asking a different question.
@tim_keary at Observer looks at what Bezos is actually building with Prometheus, and why physical-world data changes the AI moat conversation entirely.
The model race is about who has the best internet-trained AI. Prometheus is a different question: what happens when you compress the time between an idea and a manufacturable product?
A $41 billion bet with 150 people and no product shipped yet. The valuation reflects founder pedigree and supply chain access, not demonstrated output.
Blue Origin and Amazon are not just future customers. They are the answer to the hardest problem Prometheus faces: there is no internet of manufacturing data to scrape.
Honored to be quoted alongside some sharp voices in this one.
Worth reading for the competitive context and the team Bezos has assembled.
observer.com/2026/06/jeff-be…
#PhysicalAI #Prometheus #ArtificialIntelligence #EnterpriseAI
At Info-Tech LIVE 2026 in Las Vegas last week, which brought together almost 4,000 IT professionals, the role of the CIO was one of the top topics of discussion. The central theme: how leaders can move from hype to value with agentic AI.
More broadly, across the industry, there is a recognition that CIOs not only need to lead IT strategy and drive innovation with agentic AI, but they also need to connect it to clear value, and fast.
Read more:
forbes.com/sites/timkeary/20…
Recently got the opportunity to interview May Habib, CEO and co-founder of Writer, an enterprise AI platform currently valued at $1.9 billion. In this interview, we discuss her partnership with CTO and co-founder, Waseem Alshikh, the company’s agentic AI journey, and more.
observer.com/2026/04/writer-…
The alleged Claude Mythos breach reported by Bloomberg raises new questions about the security of frontier AI models. But what are the implications for security teams?
forbes.com/sites/timkeary/20…
Is Claude Mythos just a PR stunt or the start of a new era in offensive AI? I spoke to Rob Lee, chief AI officer and chief of research at SANS Institute, and Kara Sprague, CEO of HackerOne, to find out what Mythos means for the threat landscape. Read more below:
forbes.com/sites/timkeary/20…
With Nvidia GTC just days away, here are some of the top 5 sessions to watch at this years event:
forbes.com/sites/timkeary/20…
Are AI agents safe for enterprise use?
Recently I had a chance to interview Assaf Elovic, Monday's Head of AI, John Kindervag, creator of the zero trust framework, and Yaron Kassner, co-founder and CTO of cybersecurity vendor Silverfort, to find out how businesses can reduce risk when using AI agents.
Read more here:
forbes.com/sites/timkeary/20…
Tim Keary retweeted
"Groq has rapidly become one of the most exciting providers in the LLM market and is in a solid position to be a key enabler of AI model development in the future." -- @tim_keary
hubs.la/Q02vNnSj0
New research from PagerDuty finds that 98% of surveyed Fortune 1000 executives are directing their teams to pause generative AI initiatives until they establish usage policies.
@pagerduty
#GenerativeAI #AI #LLM
businesswire.com/news/home/2…
Tim Keary retweeted
ChatGPT Launches the GPT Store: OpenAI’s Answer to the App Store? techopedia.com/chatgpt-launc… via @tim_keary
The WGA's new contract with key Hollywood studios demonstrates how workers can demand a seat at the table to determine the future of AI adoption.
techopedia.com/hollywood-str…
@WGAWest
#WGAStrike #AI #ChatGPT
The CIA's move to develop a ChatGPT-style chatbot highlights that generative AI adoption is gaining a whole new momentum.
techopedia.com/us-cias-chatb…
#CIA #ChatGPT #GenerativeAI
To mark Insider Threat Awareness Month, 8 security leaders, analysts, and experts explain how organizations can protect their systems from malicious insiders.
techopedia.com/security-expe…
#InsiderThreatAwarenessMonth #Cybersecurity #CISOs
A deep-dive into what we know about Tencent's new LLM so far.
techopedia.com/is-tencents-c…
#AI #chatgpt #generativeAI
What happens when threat actors attempt to use generative AI for bad? Here's A brief look into the strange world of underground "dark AI tools."
techopedia.com/the-future-of…
@slashnextinc
@Netenrich
#AI #generativeAI #ChatGPT
ChatGPT Enterprise has some key limitations that will likely slow adoption.
techopedia.com/chatgpt-enter…
#ChatGPT #AI #GenerativeAI
@Gartner_inc