Apple has revealed a new plan to integrate AI into its devices.
Here's the rundown of everything going on in AI right now:
- A new version of Siri
-Auto-complete in Messages
-Playlist creation in Music
-Code suggestions in Xcode, and more.
Here's the rundown of everything going on in AI right now:
- A new version of Siri
-Auto-complete in Messages
-Playlist creation in Music
-Code suggestions in Xcode, and more.
TechCrunch
Apple’s job listings suggest it plans to infuse AI in multiple products
Apple seems to be finally getting serious about infusing generative AI into its products after announcing a solitary “Transformer” model-based autocorrect feature on iOS 17 earlier this year.
ULTRA - a pre-trained foundation model for knowledge graph reasoning that works on _any_ graph and outperforms supervised SOTA models on 50+ graphs.
Arxiv: arxiv.org/abs/2310.04562
GitHub here.
Arxiv: arxiv.org/abs/2310.04562
GitHub here.
A long-awaited era in Apple’s health history is here
The Apple Watch is now being leveraged to help manage Parkinson's disease, a degenerative brain disorder that affects as many as a million Americans.
Over the past year, the FDA has cleared three Apple Watch apps from developers to track the tremors and dyskinesia associated with Parkinson’s — work that began nearly a decade ago.
The goal is to help inform treatment decisions for people and providers, but there’s a long slog ahead before doctors and the broader health system can be convinced the tools are worth using.
The Apple Watch is now being leveraged to help manage Parkinson's disease, a degenerative brain disorder that affects as many as a million Americans.
Over the past year, the FDA has cleared three Apple Watch apps from developers to track the tremors and dyskinesia associated with Parkinson’s — work that began nearly a decade ago.
The goal is to help inform treatment decisions for people and providers, but there’s a long slog ahead before doctors and the broader health system can be convinced the tools are worth using.
STAT
For Parkinson’s disease, advances spurred by Apple Watch offer a glimmer of hope
After nearly a decade of development, Apple Watch is now being leveraged on an entirely new health frontier: Parkinson's disease
🔥3
Butterfly and Forest Neurotech announced collaboration to develop the first implanted, whole-brain neural interface
Deal includes $20MM in milestones paid to Butterfly with additional revenue anticipated upon commercialization for every unit sold.
Collaboration highlights one of the many potential applications of Butterfly’s novel Ultrasound-on-Chip technology for complementary technology development and device enablement.
Deal includes $20MM in milestones paid to Butterfly with additional revenue anticipated upon commercialization for every unit sold.
Collaboration highlights one of the many potential applications of Butterfly’s novel Ultrasound-on-Chip technology for complementary technology development and device enablement.
BusinessWire
Butterfly Network Announces 5-Year Co-Development Agreement with Forest Neurotech for Next-generation Brain Computer Interfaces…
Butterfly Network, Inc. (“Butterfly”) (NYSE:BFLY), a digital health company transforming care through the power of portable, semiconductor-based ultra
🔥China's Generative AI Market Map
A parallel universe to the West & an emerging "generative AI arms race"
Open AI = Zhipu AI
- Zhipu AI raised $341M this year from tech giants Alibaba & Tencent
- Released a ChatGPT-like product in August
China lags behind the West in model size and performance.
A parallel universe to the West & an emerging "generative AI arms race"
Open AI = Zhipu AI
- Zhipu AI raised $341M this year from tech giants Alibaba & Tencent
- Released a ChatGPT-like product in August
China lags behind the West in model size and performance.
❤3
⚡️Big leak: Google's secret AI tool codenamed 'Stubbs' was just exposed
According to a leak by Bedros Pamboukian on Medium, Google is set to reveal a new AI tool called 'Stubbs' on Google Makersuite.
Stubbs can generate, deploy, and even publish AI apps with no-code.
He shared screenshots to provide validity to his claim of exclusive access.
The leaks also came with news for Gemini:
-Gemini will replace Google’s PaLM-2 and may power Google's Makersuite and Vertex AI.
-Demis Hassabis said Gemini will surpass the capabilities of GPT-4
-Speculation of a late 2023 release date.
According to a leak by Bedros Pamboukian on Medium, Google is set to reveal a new AI tool called 'Stubbs' on Google Makersuite.
Stubbs can generate, deploy, and even publish AI apps with no-code.
He shared screenshots to provide validity to his claim of exclusive access.
The leaks also came with news for Gemini:
-Gemini will replace Google’s PaLM-2 and may power Google's Makersuite and Vertex AI.
-Demis Hassabis said Gemini will surpass the capabilities of GPT-4
-Speculation of a late 2023 release date.
👍4
Driving_Creativity_and_Innovation_through_EI_1698237796.pdf
552.1 KB
This paper delves into the relationship between Emotional Intelligence (EI) and innovation in organizations.
It systematically reviews existing literature to explore how EI, encompassing self-awareness, self-control, empathy, and social skills, can drive creativity and innovation.
It also highlights research gaps, proposing directions for future studies to further understand how organizations can leverage EI to foster a creative and innovative environment.
1. The paper emphasizes the evolving needs of customers and society, necessitating organizations to be emotionally intelligent to adapt their strategies and remain relevant. EI helps in understanding and managing the emotions of internal and external stakeholders, improving the quality of innovations.
2. It discusses different models and definitions of EI, such as the ability model, trait model, and mixed model, each providing a unique perspective on how individuals can understand, manage, and utilize emotions effectively in organizational contexts.
3. The paper identifies a gap in the literature regarding a systematic review that converges the fields of EI and creativity and innovation, motivating the current study. It aims to help organizations identify strategies to build EI competencies that promote innovation and creativity.
4. The study also explores the negative aspects of EI, such as higher stress levels in emotionally intelligent people and a tendency to avoid risks and maintain the status quo, which might be counterproductive to innovation.
It systematically reviews existing literature to explore how EI, encompassing self-awareness, self-control, empathy, and social skills, can drive creativity and innovation.
It also highlights research gaps, proposing directions for future studies to further understand how organizations can leverage EI to foster a creative and innovative environment.
1. The paper emphasizes the evolving needs of customers and society, necessitating organizations to be emotionally intelligent to adapt their strategies and remain relevant. EI helps in understanding and managing the emotions of internal and external stakeholders, improving the quality of innovations.
2. It discusses different models and definitions of EI, such as the ability model, trait model, and mixed model, each providing a unique perspective on how individuals can understand, manage, and utilize emotions effectively in organizational contexts.
3. The paper identifies a gap in the literature regarding a systematic review that converges the fields of EI and creativity and innovation, motivating the current study. It aims to help organizations identify strategies to build EI competencies that promote innovation and creativity.
4. The study also explores the negative aspects of EI, such as higher stress levels in emotionally intelligent people and a tendency to avoid risks and maintain the status quo, which might be counterproductive to innovation.
🔥3
AI search engine YOU launched AI "Smart Personalization" to tailor responses based on user’s interests.
Essentially, it learns to write a better prompt for YOU over time without using sensitive data like traditional search engines.
Essentially, it learns to write a better prompt for YOU over time without using sensitive data like traditional search engines.
A standard neural net optimized for compositional skills, can mimic human systematic generalization (SG) in a head-to-head comparison.
SG is the algebraic ability to understand and produce novel combinations from known components. For instance, once a child learns how to "skip", they understand how to "skip twice" or "skip around an obstacle."
SG is the algebraic ability to understand and produce novel combinations from known components. For instance, once a child learns how to "skip", they understand how to "skip twice" or "skip around an obstacle."
Nature
Human-like systematic generalization through a meta-learning neural network
Nature - The meta-learning for compositionality approach achieves the systematicity and flexibility needed for human-like generalization.
❤4
Quantum_Computing_Use_Cases__1698324460.pdf
335 KB
Quantum Computing in Financial Services. A short report from IBM with an emerging list of usecases which are relevant for the financial services industry
The financial services industry is driven by data, intricate algorithms, and rapid decision-making, making it a prime candidate for the adoption of quantum computing.
QC has the potential to impact various parts of financial services including and not limited to Portfolio optimization, Fraud detection, Cybersecurity, Risk management, Algorithmic trading, Financial modeling.
The financial services industry is driven by data, intricate algorithms, and rapid decision-making, making it a prime candidate for the adoption of quantum computing.
QC has the potential to impact various parts of financial services including and not limited to Portfolio optimization, Fraud detection, Cybersecurity, Risk management, Algorithmic trading, Financial modeling.
👍3❤2
And for those looking for evidence that being together in person matters: "Neural signaling during online exchanges was substantially suppressed compared to activity observed in those having face-to-face conversations, researchers found."
News-Medical
Zoom conversations suppress brain activity, Yale study finds
When Yale neuroscientist Joy Hirsch used sophisticated imaging tools to track in real time the brain activity of two people engaged in conversation, she discovered an intricate choreography of neural activity in areas of the brain that govern social interactions.
🤝2
Accuracy_and_Reliability_of_Chatbot_Responses_to_Physician_1698409120.pdf
956.2 KB
The accuracy and reliability of responses generated by a chatbot, specifically ChatGPT, to medical queries posed by physicians
The study involved 33 physicians across 17 specialties who generated 284 medical questions, which were then answered by the chatbot. The physicians evaluated the chatbot’s responses based on accuracy and completeness. The study aimed to assess the potential of natural language processing tools like chatbots in enhancing the accessibility of medical information for health professionals and patients, and their suitability in clinical settings.
1. The chatbot predominantly generated accurate information in response to diverse medical queries as judged by academic physician specialists. The median accuracy score was 5.5 on a scale of 1 to 6, where 6 indicates a completely correct response.
2. The chatbot’s responses were found to be comprehensive, with a median completeness score of 3 on a scale of 1 to 3, where 3 indicates a complete and comprehensive response.
3. The study found that the chatbot’s accuracy improved over time and between two different versions of the chatbot (GPT-3.5 and GPT-4).
4. The chatbot showed substantial improvement when re-queried with questions that initially received inaccurate answers.
5. The study also revealed that the chatbot had limitations and there were areas where further research and model development were needed to correct inaccuracies.
The study involved 33 physicians across 17 specialties who generated 284 medical questions, which were then answered by the chatbot. The physicians evaluated the chatbot’s responses based on accuracy and completeness. The study aimed to assess the potential of natural language processing tools like chatbots in enhancing the accessibility of medical information for health professionals and patients, and their suitability in clinical settings.
1. The chatbot predominantly generated accurate information in response to diverse medical queries as judged by academic physician specialists. The median accuracy score was 5.5 on a scale of 1 to 6, where 6 indicates a completely correct response.
2. The chatbot’s responses were found to be comprehensive, with a median completeness score of 3 on a scale of 1 to 3, where 3 indicates a complete and comprehensive response.
3. The study found that the chatbot’s accuracy improved over time and between two different versions of the chatbot (GPT-3.5 and GPT-4).
4. The chatbot showed substantial improvement when re-queried with questions that initially received inaccurate answers.
5. The study also revealed that the chatbot had limitations and there were areas where further research and model development were needed to correct inaccuracies.
❤2👍1🥴1
Many startups just died. Because OpenAI added PDF chat. You can also chat with data files and other document types.
Wrappers are being squeezed by OpenAI on one side and incumbents on the other.
It's a rough world out there.
Wrappers are being squeezed by OpenAI on one side and incumbents on the other.
It's a rough world out there.
🔥6
Animoca Brands has announced the acquisition of Web3 streaming platform Azarus for an undisclosed price. Azarus allows users to stream gameplay and interact with viewers in real time, leveraging blockchain-based tokens for incentives.
VentureBeat
Animoca Brands acquires Azarus to bring streaming to Web3 games
In a quest to expand metaverse gaming, Animoca Brands has announced the acquisition of Web3 streaming platform Azarus.
Drake_Star_Global_Gaming_Report_Q3_2023_1698668012.pdf
4.2 MB
The Drake Star Q3'23 Global Gaming Report is out
Q3 saw gaming M&A activity from some of the biggest strategics and Q4 started on a great note with the era-defining gaming deal in Microsoft / Activision finally closing.
Highlights include:
1. Tencent leads the way with five deals during the quarter including the Techland acquisition which is among the top 3 largest deals of the year. Other notable buyers were Playtika (Innplay Labs and YoudaGames), Goldman Sachs, General Atlantic and the LEGO Group (Kahoot!), Take-Two Interactive, Behaviour Interactive, Capcom and Roblox.
2. $1B in 185 private financing rounds, an increase in value from Q2'23. Notable raises were Candivore ($100M), Second Dinner ($90M), Story Protocol ($54M), Futureverse ($54M) and Inworld AI ($50M).
3. BITKRAFT Ventures led the VC league table, followed by Andreessen Horowitz, Play Ventures, Griffin Gaming Partners and vgames.
4. Tencent, Sony, Take-Two Interactive and Savvy Games Group / Scopely are expected to be the most active buyers.
Q3 saw gaming M&A activity from some of the biggest strategics and Q4 started on a great note with the era-defining gaming deal in Microsoft / Activision finally closing.
Highlights include:
1. Tencent leads the way with five deals during the quarter including the Techland acquisition which is among the top 3 largest deals of the year. Other notable buyers were Playtika (Innplay Labs and YoudaGames), Goldman Sachs, General Atlantic and the LEGO Group (Kahoot!), Take-Two Interactive, Behaviour Interactive, Capcom and Roblox.
2. $1B in 185 private financing rounds, an increase in value from Q2'23. Notable raises were Candivore ($100M), Second Dinner ($90M), Story Protocol ($54M), Futureverse ($54M) and Inworld AI ($50M).
3. BITKRAFT Ventures led the VC league table, followed by Andreessen Horowitz, Play Ventures, Griffin Gaming Partners and vgames.
4. Tencent, Sony, Take-Two Interactive and Savvy Games Group / Scopely are expected to be the most active buyers.
AI and biotech lead in bigger week for funding deals
3 other U.S. startups last week raised nine-figure rounds and there were several other big rounds just under that.
Biotech seemed to be the big winner, taking three of the top five spots.
3 other U.S. startups last week raised nine-figure rounds and there were several other big rounds just under that.
Biotech seemed to be the big winner, taking three of the top five spots.
Crunchbase News
The Week’s 10 Biggest Funding Rounds: Anthropic Rules With $500M Round; Aiolos Bio Nabs Large Raise
This week saw a nice pickup thanks again to AI. Four companies raised nine-figure rounds and there were several other big rounds just under that. Of course, Anthropic’s round late in the week was the top raise, but biotech took three of the top five spots.
Microsoft might've just leaked the "secretive" parameter count of GPT 3.5 in their CodeFusion paper.
Achieving this quality of chat with 20B parameter is very impressive.
Achieving this quality of chat with 20B parameter is very impressive.
Qualcomm's brand new Snapdragons - Snapdragon 8 gen 3 - support 10B LLMs on smartphones for the future of genAI, which is multi-modal.
EE Times
Snapdragon 8 Gen 3 Evolves Smartphones to Genius-Phones - EE Times
Thanks to Qualcomm's latest Snapdragon 8 Gen 3 mobile SoC, multi-modal generative AI will be coming soon to smartphones.