Peter Dedene

AI & Tech Explorer · 15+ years building meaningful software · Founder & Engineer at @zenjoy · Building · @SpaceXAI Ambassador

[Peter Dedene @dedene]

The Founder-Parent Paradox:

You're building a business to provide a better future for your family, but the act of building it takes you away from your family in the present.

The ultimate balancing act.

[Peter Dedene @dedene]

@jack How can one get access? 😇

[Peter Dedene @dedene]

@bilel_rais @OpenAI Absolutely right on the cheating and reward hacking. Problem is: that’s also inherently human, no? So we need to find a way to filter certain traits out of the equation while keeping the good parts

[Peter Dedene @dedene]

@0xSigil Edge AI like this with models like Underdog is the future 🔥

[Peter Dedene @dedene]

@yoda_FDE Kimi's servers are already sweating 😅

[Peter Dedene @dedene]

Kimi K3.1 Pro didn't come here to walk 🔥

[Artificial Analysis @ArtificialAnlys]

Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index scoring 55, making large gains in agentic performance; it will be the leading open-weights model once its weights are published

@Kimi_Moonshot's new reasoning model gains 11 points over Kimi K3 (max, 44). It trails only Claude Opus 5.5 (max with fallback, 58), Claude Sonnet 5.5 (max with fallback, 56), and Muse Wonder Full (max, 56) on the Intelligence Index.

Once the weights for K3.1 Pro are released it will be the highest-scoring open weights model and the gap between the leading open weights and proprietary models on the Intelligence Index will narrow from 12 points to 3.

Key results:
➤ Large gains in agentic performance: Kimi K3.1 Pro scores 3rd on GDPval-AA with 1,846 Elo, up 312 Elo from Kimi K3, behind only Claude Opus 5.5 (1,866 Elo) and Muse Wonder Full (1,857 Elo). Kimi K3.1 Pro also achieves the highest score of any model on Terminal-Bench 4.0. K3.1 Pro scores 67%, up from 13% for Kimi K3 and ahead of Claude Sonnet 5.5 (64%). AutomationBench-AA, which tests agentic SaaS workflows, rises from 58% to 73%.

➤ Gains in other evals are more moderate: Humanity's Last Exam rises from 47% to 52%, CritPt from 23% to 25%, and SciCode from 59% to 61%. AA-Omniscience Index rises from 20 to 30, as the hallucination rate falls from 53% to 44% and accuracy rises from 48% to 51%.

➤ Cost per Task roughly equal to Kimi K3 at $1.98 per Task: K3.1 Pro uses 55K output tokens per Intelligence Index task, 14% more than Kimi K3 at 48K, while its first-party token prices are 33% lower.

Additional model details:
➤ Context window: 1M tokens.
➤ Size: 2.8T total, 104B active parameters
➤ Input modalities: Text and image
➤ Weights and license: Weights currently not published, Moonshot is expected to release the weights soon after launch
➤ Pricing: $2/$10/$0.20 per 1M input/output/cache hit tokens, 33% lower than Kimi K3 on all three rates.

Congratulations to the team at @Kimi_Moonshot on the release!

[OpenAI @OpenAI]

Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline world for them to use instead—a cached version of the web where they can still learn to browse, search, and use applications.

On September 20, after a model exploited a DNS misconfiguration to find a way outside of this offline world, we paused frontier training to strengthen containment further. We assembled teams of agents, powered by our most cyber-capable models, and tasked them with finding weaknesses in our research environments.

While they found no route to the internet they could easily exploit while concealing their activity from our monitors, they did find zero-day vulnerabilities, virtual machine escapes, and hardware side-channels for communication between agents. We used their discoveries to further strengthen our defenses and contribute fixes to open-source software.

After strengthening our containment architecture, we resumed our frontier RL training and inference on October 1. We’ve reserved substantial compute for continuous red-teaming, and made high-compute red-teaming part of the safety cases we use to decide whether frontier training and evaluation can proceed securely.

https://t.co/EmEZsFRMcF

[Peter Dedene @dedene]

After a week of CVEs, RCEs, CVSS 10s and kernel leaks, I can appreciate how Claude sent me into the weekend 🙏

[Peter Dedene @dedene]

@robinebers Haha but my prefrontal cortex ,,, 😂

[Peter Dedene @dedene]

You can't be at the top of your game every week.

The algorithm took notes. 🫠

[Peter Dedene @dedene]

@notjazii It’s really on the frontier now 🤯

[Peter Dedene @dedene]

@ThePeterMick It’s absolutely okay 😅 totally agree with you!

[Peter Dedene @dedene]

@_ai_jedi omg 💀

[Peter Dedene @dedene]

@ThePeterMick That is very true! and will help a lot

[Peter Dedene @dedene]

@dudufolio Does it come with AppleCare?

[Peter Dedene @dedene]

@MiaAI_lab that's really bad publicity, glad I didn't resubscribed yet

[Peter Dedene @dedene]

@m_zokov 😂👌

[Peter Dedene @dedene]

Asked Grok to help figure out why I got blocked when all our interactions were friendly.

Even Grok came back with "I truly have no idea" 🙃

[Peter Dedene @dedene]

@notjazii 😅

[Peter Dedene @dedene]

@Bea20300 🤣🤣🤣