逼真且富有想像力
OpenAI, the maker of chatbot ChatGPT and image generator DALL-E, unveiled its newest artificial intelligence model Sora on February 15.
創造ChatGPT聊天機器人和圖像生成器DALL-E的OpenAI公司，於2月15日發布了最新的人工智慧模型Sora。
chatbot(n.)聊天機器人
As the company’s newest entry into the generative AI market, Sora is a “text-to-video model that can generate videos up to a minute long with impressive quality and detail,” according to ZDNet’s Sabrina Ortiz.
根據商業技術新聞網站ZDNet的莎賓娜·歐提茲報導，Sora是OpenAI進入生成式人工智慧市場的最新產品，是一個「文字轉影片模型，可以生成長達一分鐘的影片，其品質和細節令人印象深刻」。
entry(n.)(進入某市場的)產品、參賽作品generate(v.)產生;生成generative(adj.)生成式的
While Sora is not yet available to the public, OpenAI posted a link on X to its website, which offers various examples of the model’s “realistic and imaginative” video-making capabilities.
雖然Sora尚未對大眾開放，但OpenAI在社群媒體X上發布了其網站連結，網站提供了該模型「逼真且富有想像力」的影片生成能力的各種範例。
realistic(adj.)寫實的;逼真的=lifelikecapability(n.)能力=ability
On the site, OpenAI states that it is teaching AI to “simulate the physical world in motion” as a way to aid people in solving problems that “require real-world interaction.”
OpenAI在網站上表示，正在教導AI「模擬運動中的物理世界」，以幫助人們解決「需要真實世界互動」的問題。
simulate(v.)類比、模擬、模仿simulated exam模擬考=mock exammotion(n.)運動;移動in motion運動、移動中的aid sb. in V-ing輔助某人做某事=assist sb. in V-ing
Therefore, Sora has been developed as a model that can not only create complex scenes involving “multiple characters, specific types of motion, and great detail,” but also offer a deeper understanding of language, user input prompts and real-world subjects.
因此，Sora被開發成一個不僅可以創建涉及「多重角色、特定類型的運動和大量細節」的複雜場景的模型，還能更深入地理解語言、用戶輸入指令和真實世界的主題。
not only…but also…不僅…而且…involving=which involve涉及、包含prompt(n.)指令
Ortiz notes in her initial impressions of Sora that while the model is not the first of its kind, the continuity and fluidity of its output is far superior to that of other models, such as Runway AI’s text-to-video model, Gen-2.
歐提茲提到她對Sora的初步印象，雖然該模型並非首創，但其輸出的連貫性和流暢度遠優於其他模型，如Runway AI公司的文字轉影片模型Gen-2。
initial(adj.)最初的=originalthe first of its kind首創continuity(n.)連貫性fluidity(n.)流暢度output(n.)輸出input(n.)輸入superior(adj.)好於平均的、好過其他的、優越的that=the continuity and fluidity of the output
However, although Sora has created impressive videos, it is not without weaknesses.
不過，儘管Sora創作的影片令人印象深刻，但它並非沒有缺點。
weakness(n.)缺點=drawback=disadvantage=demeritstrength(n.)長處;優點=benefit=advantage=merit
Among those outlined, the model may struggle to simulate complex scenes, may not fully understand cause and effect and “may also confuse spatial details of a prompt.”
其中，Sora可能難以模擬複雜的場景，可能無法完全理解因果關係，「還可能混淆指令的空間細節」。
Among those outlined=Among those drawbacks that have been outlinedstruggle to VR做某事時遭遇困難=have difficulty V-ingcause and effect因果關係space(n.)空間spatial(adj.)空間的spatial awareness空間概念
According to OpenAI, Sora is already becoming available to “red teamers” that will test the safety and potential risks of the model by trying “to make the platform fail, produce inappropriate content, or go rogue.”
根據OpenAI的說法，Sora已經開始提供給「紅隊演練團隊成員」，這些成員將透過嘗試「使平台當機、生成不當內容或變得失控」來測試模型的安全性和潛在風險。
red teamers紅隊演練團隊成員inappropriate(adj.)不適合的;不恰當的appropriate(adj.)恰當的;合宜的go rogue擅自行動、行為反常、不受控=go out of order
The company has also noted that it will start engaging global policymakers, educators and artists in dialogues about their concerns over the new AI model.
OpenAI還指出，將開始邀請全球政策制定者、教育工作者和藝術家加入對話，了解他們對這一新人工智慧模型的顧慮。
engage sb. in sth.使～參與、投入～之中
DW News has reported that questions have already been raised regarding the “ethical and societal implications” of Sora’s high-quality videos as well as the sources used to train the model.
《德國之聲》報導，已經有人對Sora高品質的影片產生的「道德和社會影響」以及訓練模型的資料來源提出了質疑。
raise questions(正式)提出問題、質疑regarding(prep.)關於implication(n.)可能的影響、可能的後果=consequence=ramification