Microsoft calls for regulation of facial recognition, saying it’s too risky to leave to tech industry alone ⏭ great wisdom but will it be followed? #FacialRecognition #Privacy
#FacialRecognition shouldn’t be everywhere and it shouldn’t look at everyone. But when the Chinese govt subsidises its #AI start-ups to compete with commercial entities, it opens the (back) door to a sinister big brother harvesting our biometric data.
#facialrecognition technology presents risks to privacy. @TRSMRyersonU 's @AnnCavoukian of the Privacy by Design Center of Excellence, describes how this tech can be misused.
The police have been rolling out privacy-invading, intimidating & discriminatory #FacialRecognition surveillance tech on our streets. If you're at #CarnivalOfResistance tomorrow, find us and learn how to con the cameras with face paint to #ResistFacialRec
但是,發展臉部辨識研究的美國聯邦機構國家標凖技術研究所(National Institute of Standards and Technology)的生物辨識項目負責人格羅瑟(Patrick Grother)表示,相關技術正變得越來越可靠,將原因歸功於「卷積神經網路」(Convolutional Neural Network,CNN)的發展。
到底會有多少公家機關或是私人公司秉持著「大眾利益」,行「侵害隱私」之實呢?
2017年在威爾斯卡地夫警方使用最新的人臉辨識技術從現場17萬名觀眾中找出了2470名嫌疑罪犯,但之後卻被發現當中出現「大量錯誤」。
微軟呼籲管制臉部辨識
用於臉部辨識的AI儘管方便,卻也帶來隱私疑慮(圖片取自Affectiva)
乘客在法國機場的移民局接受臉部辨識。(湯森路透)
The cameras that know if you're happy - or a threat - #MachineLearning #FacialRecognition
via
18日消息,亞馬遜臉部辨識工具Rekognition由於與美國警方合作,受到員工反對。一名員工在Medium平台匿名發表公開信,對公司與警方的合作表示擔憂。
事情起源是15日,亞馬遜CEO貝佐斯在Wired會上為美國國防部JEDI雲端合約辯護時,強調「做出可能不受歡迎的決定(如與軍方合作)是商業領袖工作的一部分」。僅過一天後,這封員工公開信就在各大媒體傳開了。