帰ってきて!
Henry Cavill on the possibility of returning as Superman following James Gunn’s departure from DC leadership:
“James Gunn is a great guy, and I genuinely respect what he has contributed to DC, but you know, everyone has their flaws. The problem is that James wasn’t always willing to acknowledge his own, and sometimes that can get in the way of something truly special.
As for Superman, my answer has always been simple: if DC calls, I’m ready to put that suit back on. And if Zack Snyder and Ben Affleck are also returning to help complete the SnyderVerse, then I don’t see any reason to say no.
I know what Superman means to millions of fans around the world. I know how much they fought to see my version of the character get the ending it deserved. I owe those fans something special, and if I get another opportunity, I intend to give them the Superman performance they’ve been waiting for all these years.
The cape was never the problem. The story was never finished. And perhaps, this time, we can finally finish what we started.”
26
わかる....
欧州駐在経験でこういうのに感化され、美しいが内容のないPPTと、MBAあたりのロジックで煙にまくJTC幹部層が増えていくのだ
Replying to @English_Victory
4年いて分かった真実
・マネージャー以上は遅くまで働く
・資料はキレイ。でも、何が言いたいか分からない
・結論がない。中身が薄い
・自信満々に話すからそれっぽく聞こえる
・TOEIC最高680点の私。分からなかったのは、英語のせいだけじゃなかった
1年目の私に言いたい。
理解できないのは、英語だけじゃないぞ!
1
84
何だこれは....
AI生成?
まるでガングリフォンじゃないか...
Humanoid robotics just crossed into a very different visual territory. 🤖⚡
This footage depicts a coordinated group of airborne humanoid machines navigating a dense forest, avoiding obstacles and transitioning from aerial movement to ground operations.
The fascinating part isn’t simply flight.
It’s the concept of combining:
🧠 Autonomous coordination
👁️ Environmental perception
⚡ High-speed navigation
🤖 Humanoid mmobility
If machines can eventually coordinate across air and ground with this level of autonomy, robotics stops being about individual robots—and starts becoming about intelligent robotic systems working as a team.
The implications would be enormous.
#Robotics #ArtificialIntelligence #AI #IoT #5G #NLP #Technology #Tech #innovation #AutonomousSystems #FutureTech
Cc: @jblefevre60 @Nicochan33 @pierrepinna @RosyCoaching @kalydeoo @Ym78200 @enilev @FrRonconii @CurieuxExplorer @Khulood_Almani @mvollmer1 @JoannMoretti @margaretsiegien @Shi4Tech @anand_narang @Eli_Krumova @HaroldSinnott @bimedotcom @EvanKirstel @Hana_ElSayyed @sallyeaves @SpirosMargaris @ipfconline1 @LaurentAlaus @Analytics_699 @Jeffkagan @AndresVilarino @efipm @FMFrancoise @MHcommunicate @Corix_JC @JolaBurnett @JeromeMONANGE @mary_gambara @drsharwood @smoothsale @sulefati7 @debashis_dutta @dinisguarda @devbhatt72 @terence_mills @BetaMoroney @mdrechsler @MarshaCollier @avrohomg @NevilleGaunt @PamMktgNut @arlenenewbigg
Readers added context they thought people might want to know
The video is not real footage of humanoid robots. No such "AIRO" coordinated airborne humanoids exist; AIRO develops drones and aircraft while real jet-powered humanoids remain limited single-unit research prototypes.
investor.theairogroup.com
arxiv.org/abs/2506.01125
forbes.com/sites/leslieka…
eurekalert.org/multimedia/107…
58
うめ長 retweeted
A clip of the Godzilla drone show at Dodger Stadium. We’ll have the full video on our Youtube shorty after it’s over.
素晴らしい内容。
フィジカルAIに進むと、多分、こういう思考になる。我々も現実世界の課題にぶつかると、このような事を考える。
こんなポストが言語の壁を超えて読めるようにしてくれたイーロンには感謝しかない
我终于明白为什么LeCun顶着全网群嘲也要死磕世界模型了,
现在很多人误以为大模型快要通向AGI了,纯粹是因为被那种博导般的答题体感给骗了,
@ylecun 说单靠把大语言模型做大,哪怕到了地狱也搞不出人类水平智能,
甚至直接开喷数据中心养出天才军团纯属扯淡,很多人觉得他迂腐、天天唱衰,
但我仔细翻完他在ECCV上的最新拆解才彻底明白,他反对的从来不是语言模型有用,他自己做了一辈子模型太清楚了,大模型说白了就是把全人类写过的几千亿字当成了超级开卷题库,
你问什么它都能对号入座吐出来,这种博导体感是工程的巨大成功,但也让很多人误以为它真的学会了独立思考
有意思的是LeCun搬出了皮亚杰那句老话,智能从来不看你知道什么、看的是你不知道的时候该怎么做,拿学开车来说,
人类小孩四岁光靠眼睛看世界就吸进几十万亿字节,二十小时练习就能上路,而我们的大模型相当于读了人类四十万年的书,
在物理世界里连个洗碗叠衣服的家务机器人都撑不起来,换个没见过的环境分分钟崩溃,
说白了自回归只是在一个词一个词往下猜,现在的所谓推理其实大量是在词语空间里做多轮抽样搜索,遇到代码和数学这种有编译器自动给奖励的封闭沙盒还能刷分,一旦扔进没有自动打分器的真实世界,错误就会指数级放大
说实话这就是为什么他宁可离开Meta、跑到巴黎去给AMI Labs融三十五亿美元估值,也要死磕JEPA和世界模型,他要的是让机器学会预测我做这件事世界会发生什么,
在抽象空间里做动作规划,不用在文字表面拼漂亮句子,连现在的缩放派其实都在偷偷向他妥协,单独拉视觉编码器、在隐空间里做搜索、用物理引擎搞模拟,
外壳虽然还叫LLM,内核其实都在往世界模型的方向打补丁,大家争论到最后已经不是要不要算力,而是我们要把算力花在背诵更多文字上,还是花在让机器看懂物理因果上
我自己平时也天天靠AI写代码查资料,缩放派确实把已知世界的说明书背到了极致,但真正的智能是要去写说明书从来没写过的那一页,
以后别再看各家发布会画什么天才之国的饼,就盯紧三件事,新任务要多少样本能学会,换个环境会不会整段崩溃,
以及汽车和机器人能不能像人一样在物理世界里安全跑起来,谁对谁错不用在网上打嘴仗,现实世界自会交卷。
61













