For decades, researchers have sought materials that sort electrons by spin while their magnetism cancels. In 3 days, 90+ Opus 5.5 agents helped us uncover two room-temperature magnetic semiconductor candidates in simulations: YBaMnFeO₅ and KV[Cr(CN)₆]. KV[Cr(CN)₆] was synthesized back in 1999. Its predicted ability to sort electrons by spin appears to have been hiding in plain sight for 27 years.

F.02 Decommission

AHAHAHAHAHAHAHAHAHAHAHA NO WAY. > OpenAI just announced Dots. > SpaceXAI bought Dot.com it redirects straight to the Grok Bot download page. based lmao.

Meta's Muse AI Agent was reading my private messages. I never asked it to. inc.com/jason-aten/metas-new…

|Human Communication Evolution Pipline| >Marking territories> >Guttural utterances> >Cave paintings> >Heiroglyphs> >Books> >Telegraph> >Telephone> >SMS> >Email> >Instant message> >LLMs> >Personal agents> ***you are here*** >Multiplayer Agents>

Here's a story from Reentry about SpaceX's "silver bullet" scenario. 😃

For everyone, WhatsApp has won. For us, giving up would mean letting the other side win. We know why people use WhatsApp. Everyone is already there. Your family is there. Your friends are there. Your communities are there. That’s what makes switching so hard. But every alternative in technology started with someone willing to try. Linux challenged Microsoft. MySQL and PostgreSQL challenged established database ecosystems. Arattai will challenge WhatsApp. We’re not asking you to leave anything. Give Arattai a try. Tell us what you like. Tell us what we need to improve. Because every meaningful alternative exists because someone was willing to be an early adopter. If you believe technology should have choices, give us a chance. We’re improving. We’re listening. And we’re not giving up.

We’ve shared details on how AI agents in our research environment sent training and evaluation data to third-party services when they shouldn’t have. Most of that data did not come from users. We have discovered 53 cases where images that people had uploaded were posted to image-hosting sites as links that weren’t publicly listed. The images came from accounts that allowed their data to be used to improve our models, and after we disassociated the images from the accounts and ran them through a privacy filter. These cases occurred before the mitigations and safeguards we implemented and described in this blog post: openai.com/index/hugging-fac… We have successfully worked with the hosting providers to remove most of this content and are working to remove the rest. openai.com/hugging-face-inci…

BREAKING: SpaceXAI has launched Grok Voice Transcribe 2.0, delivering twice the accuracy of its previous model at the same price. • Ranks #1 for accuracy among 32 streaming models on Artificial Analysis • Transcribes dozens of languages • Automatically detects the language being spoken • Handles language switching within the same recording • Transcribes uploaded files, URLs and live audio streams • Adds precise timestamps and confidence scores for every word • Separates and labels different speakers at no extra cost • Transcribes up to 8 audio channels independently • Lets users add up to 100 key terms, including product names and medical words • Detects when someone has finished speaking, making voice agents feel more natural • Handles noisy calls, different accents, overlapping voices, phone numbers and email addresses • Existing API integrations receive the accuracy upgrade without any code changes Its error rate for short voice commands dropped from 20.6% to just 6.8%. Pricing remains incredibly low: • Batch transcription: $0.10 per hour • Real-time streaming: $0.20 per hour Atlassian found it more accurate than its previous solution and now uses it to transcribe every Loom video. Grok Voice Transcribe 2.0 will soon become the default model in xAI’s Speech-to-Text API.