Manipulation can leave you questioning decisions you felt certain about before the conversation started. Guilt gets added, details get twisted, pressure increases, and suddenly you’re agreeing to something you never wanted. That confusion works in their favor when someone benefits from keeping you unsure of yourself. Healthy communication gives you room to think, ask questions, disagree, and make your own choice.

Price manipulation now happens almost every week, with $LAB by @LABtrade_ becoming the latest pump-and-dump token while Bitget continues playing the usual CEX role. The LAB team, @vsadkovv, appears to control a significant portion of the supply. Wallets linked to the team still hold large allocations: 0x7Cfd8d2d8626B287bEA569b5e65AB5CBb75E9265 0x78a79D0fa0Eaf58741f5Bde7E05b5CC8F33D24d3 0x36FC85Ec486C254c9564d66de8c4210a1A20C291 0xf79ff8a5052E969a6d13E18c4E439fE5202B02Fa 0xB4b74D63F30076870d54aB9E8E6a7D18293273c3 0xe03722dedBf090Ad7A1C8F82ceB86637053E21dd On April 8, a wallet linked to the team (0xe037) deposited 40M LAB worth $13.6M to Bitget: 0x77156a0a621d2Ac7A075C0AC3172707C2e4aa191 The LAB price started pumping on May 1, but a week earlier wallets linked to the team deposited 96M LAB worth around $63M to Bitget: 0xDd77BFbDc11Cd37fD255AE35A4ac39Df1F9d570a 0x6593aa6c31C88397c37f71259625EC92Fe4EE0bF This looks coordinated. Gas fees (0.14 BNB) were distributed a week earlier 0x50f2760fd5E6d546EE7dcEB617F33497A3C38593 0x0559694BbB47dbA8Bc3B7ac93004EF401F2da16d The wallet below has also been aggressively buying $LAB on-chain and depositing to Gate and Bitget, including tokens like $SkyAI, which surged 1000% in the last 30 days: 0x11fc12b988933966688d33B70651B5f2f450963C It has been weeks since @GracyBitget promised an investigation, yet there has been no public update. If platforms cannot conduct internal investigations, identify coordinated manipulation, or provide transparency on who is behind these activities, confidence in market integrity keeps declining. ZachXBT still has a reward open for credible intel that lead to identifying the actors behind this operation. Stay smart.

A Bimanual Robot for Dynamic Manipulation AthenaZero just made the September cover of Science Robotics. {📌 Worth saving. I linked the full RAI Institute blog further down in the post if you want to dive deeper into AthenaZero’s design and control approach.} Researchers at the @rai_inst built a low-inertia bimanual robot to study one of the hardest problems in robotics: dynamic manipulation. And the first demos are pretty wild. AthenaZero can throw, catch and bat a baseball at speeds approaching human performance. • Throwing: up to 113 km/h • Catching: up to 66 km/h • Batting: around 50 km/h • Catching and batting over just 7.3 meters That short distance puts serious pressure on reaction times. But the interesting part isn't baseball. Most robot arms use high gear ratios that make them strong and precise, but also stiff and difficult to move compliantly during contact. AthenaZero takes a very different approach. Its arms use low gear ratios, as low as 5:1 in most joints, and have an effective mass at the wrist of only 3.97 kg. That's much closer to a human arm than a conventional collaborative robot. The result is a robot that can accelerate quickly, react to changing trajectories and absorb contact instead of simply fighting against it. The researchers demonstrated robot-to-robot and human-to-robot catching and batting, with the robot continuously adapting its motion to the incoming ball. Dynamic manipulation remains one of the hardest open problems in robotics. But systems like AthenaZero show what becomes possible when robots are designed for fast, physical interaction from the beginning. Blog: —— Weekly robotics and AI insights. Subscribe free:

A video world model for robot manipulation that actually verifies whether the generated video faithfully follows the prescribed actions has arrived. DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation Built on Wan2.2-TI2V-5B, this model predicts future frames with high fidelity from an initial observation, language instruction, and bimanual action trajectory. It ranked 1st among 31 teams on WorldArena 2.0 Track 1 (EWMScore-P: 60.65). 🔷 Highlight 1: PRoPE — Injecting SE(3) Directly into Attention Instead of compressing actions into generic tokens, the model injects end-effector positions, rotation matrices, and gripper states as SE(3) transformations directly into the attention mechanism. Attention heads are partitioned per arm, and token-wise transforms are applied via Kronecker products to eliminate dependence on global coordinate frames. This yields a remarkable controllability score of 98.55. 🔶 Highlight 2: Depth Branch + Object-Centric Supervision Beyond Visual Plausibility A two-pronged approach tackles the fundamental problem that RGB loss alone cannot constrain surface ordering or object extent. A lightweight depth branch (the final M blocks replicated with one-way cross-attention) enforces geometric consistency, while SAM3 masks combined with Gram matrix constraints from a frozen V-JEPA teacher preserve temporal coherence of manipulated objects — ensuring contact-local errors remain influential despite large static backgrounds. 🟣 Highlight 3: DMD Distillation Compresses Multi-Step into Few-Step Distribution-Matching Distillation (DMD) combining KL divergence and a non-saturating GAN loss drastically reduces inference steps. Trained on over 6,000 hours of diverse data spanning Ego4D, AgiBot World 2026, and RoboTwin 2.0, the model balances broad visual priors with precise action grounding. Robot world models have taken a decisive step from "looks realistic" to "moves correctly." #Robotics# #WorldModel#

AI-powered robotic dexterity is revolutionising manipulation by enabling robots to learn complex motor skills with humanlike precision, transforming industries worldwide —  @meisshaily #ArtificialIntelligence# #AI# #Robotics# #Tech# #TEchNEws# #TechNology# #Data#

🚨HEDERA EXPLOIT: SAUCE PROTOCOL HIT BY ORACLE MANIPULATION, OVER $5M STOLEN! Attacker deposited collateral, manipulated oracle prices, borrowed ~6.6M $USDC + 35M $HBAR, swapped on SaucerSwap, and bridged funds to Ethereum. Stolen assets, now ~2.36K $ETH + 15.58 $WBTC, tracked on specific Ethereum wallets. Initial funding traced to Tornado Cash.