Headlines will come and go, but Dubai will just keep getting better. Fly with us from over 140 destinations, and rediscover Dubai this winter. Customers booking tickets to Dubai across all cabins will enjoy guaranteed access to three popular excursions for travel until 13 December. Visitors can also enjoy additional value and discounts across hundreds of outlets in the city, with My Emirates Pass, Skywards, Dubai Experience and more.

lots of coding agent memory systems are unhelpful - at best i have said this a lot and some folks think I'm against agent memory in entirety which is not true - I just think that users should be on the loop that generates memories and should be able to enable / disable the system or individual memories at will and should be able to dismiss disruptive or unhelpful memories putting a user on the loop also lets you build a better data flywheel over time to improve the quality of recommendations live in @humanlayer_dev btw lmk if you want beta access

Robotics has a flywheel problem: better robots need better data, better training + better evaluation - all feeding each other. At Exploit, @quack_builder of @openroboto shows how Bittensor can connect that loop: from human demonstrations → open model competition → working robots in the real world:

If planes fly just that little bit lower, the exhaust clouds they emit could be avoided. Contrail-avoidance would leave the aviation industry better off, and the planet less polluted

Deerfield’s Jim Flynn on shrinking error bars, cutting through longevity hype, and why AI needs better loss functions. #longevity# #AI# #errorbars# #lossfunctions# #Deerfield#

NASA wants to fly Artemis III next year. But the longer NASA waits to fly Artemis III, the better chance it will have to fly with a higher-fidelity vehicle—that is, one closer to landing on the Moon than being a basic prototype.

A pigeon might be a better AI engineer than we think. Watch a pigeon eating on the street. It looks chaotic. Seeds flying everywhere. Food scattered across the ground. Constant movement. But look closer. Within seconds, it identifies exactly what it wants, ignores what it doesn't need, and picks the right food with remarkable speed and precision. Nature has been doing intelligent object recognition long before we invented AI. And there are some fascinating technology lessons here: 🔹 Computer Vision: Recognizing the right objects in a visually complex environment. 🔹 Edge AI: Making decisions instantly, without waiting for instructions from the cloud. 🔹 Selective Attention: Focusing on relevant information while filtering out distractions. 🔹 Autonomous Robotics: Identifying, selecting and picking individual objects from a messy environment. 🔹 Real-time Decision Making: Processing visual information and translating it into immediate physical action. Think about the challenge facing modern robotics: picking one specific object from hundreds of different objects, under changing conditions, at speed. A pigeon does something remarkably similar every day. Of course, biological vision and artificial neural networks work differently. But the inspiration is fascinating. Perhaps the next breakthrough in AI won't come only from building bigger models. It may come from understanding how nature achieves intelligence with extraordinary efficiency. Sometimes, the best AI laboratory is right outside your window. - Matt Gaetz