Attention, digital task jugglers!

If you're weary of the never-ending cycle of repetitive chores, look no further than Make.com. 

This innovative no-code/low-code solution is your ticket to reclaiming your precious time. With Make.com, you can effortlessly automate both personal and business-related digital tasks, unlocking unparalleled efficiency.

Say goodbye to the mundane and embrace a streamlined digital life. Don't wait; click now to experience the future of productivity with Make.com!

What happend this week at Digi-Pal?

1. Webscraping 101

2. Are you convenient enough?

3. Codeium - The Free AI-Powered Toolkit

4. How to use your voice to become more productive?

What happend this week in Tech?

1. Amazon (AWS) invests 4 Billion into Antropic

Amazon is investing up to $4 billion in Anthropic as part of a collaboration to develop reliable and high-performing AI foundation models. This partnership will make Anthropic's AI accessible to Amazon Web Services (AWS) customers. AWS will become Anthropic's primary cloud provider for mission-critical tasks, providing access to AWS Trainium and Inferentia chips for model training and deployment.

Amazon developers can build on Anthropic's models via Amazon Bedrock, enabling them to incorporate generative AI into their applications and improve customer experiences. Organizations can use Claude 2 for various tasks, including dialogue generation, content creation, reasoning, and instruction.

Several companies are already using Anthropic models on Amazon Bedrock, such as LexisNexis for legal solutions, Bridgewater Associates for financial analysis, and Lonely Planet for travel recommendations. Both Amazon and Anthropic prioritize the responsible development and deployment of AI, actively engaging with organizations like GPAI, PAI, and NIST.

Amazon will hold a minority stake in Anthropic, and their corporate governance structure remains unchanged. Pre-deployment tests will be conducted to manage AI system risks. This investment ensures Anthropic's ability to advance AI safety and research while delivering secure AI cloud technologies worldwide.

2. OpenAI now can see, hear and speak

OpenAI is introducing new voice and image capabilities to ChatGPT, offering a more interactive interface. Users will be able to engage in voice conversations with ChatGPT and share images to facilitate discussions and tasks.

Voice capabilities will be available for iOS and Android users, with five different voice options. The text-to-speech model behind this feature was created with professional voice actors, and Whisper, an open-source speech recognition system, is used to transcribe spoken words.

Image understanding is powered by multimodal GPT-3.5 and GPT-4, enabling ChatGPT to interpret a wide range of images, including photographs, screenshots, and documents containing both text and images. Users can use a drawing tool in the mobile app to focus on specific parts of the image.

OpenAI is deploying these capabilities gradually, aiming to refine risk mitigation and prepare for more advanced systems. The voice technology is initially focused on voice chat, with collaborations like Spotify's Voice Translation feature.

Vision-based models are being tested for risk in domains such as extremism and scientific proficiency before broader deployment. OpenAI is also working to ensure privacy and responsible usage, limiting the model's ability to analyze and make statements about individuals.

OpenAI is transparent about the model's limitations and encourages responsible use. Plus and Enterprise users will gain access to these features first, with plans to expand access to other user groups in the future.

3. 40 % Performance Boost by using GPT4 for knowledge workers - according to a study by BCG and Harvard School

A study led by Harvard in collaboration with Boston Consulting Group (BCG) and MIT has shown that the use of generative AI, particularly GPT-4, led to a 40% performance boost among knowledge workers, including consultants at BCG. The study involved 758 consultants and found that those using AI completed more tasks, completed them more quickly, and produced higher-quality results compared to a control group. Interestingly, the lowest-performing consultants experienced the most significant performance improvement with AI.

However, the study also highlighted the challenge of determining when AI is reliable for specific tasks, leading to the risk of overestimating AI competence. It introduced the concepts of "Centaurs" and "Cyborgs," denoting different approaches to combining human and AI work.

The study suggests that AI can lead to homogenization of ideas, and companies may need to deploy multiple AI models or increase human involvement to counteract this effect. It also emphasizes the importance of collecting clean, differentiated data for AI applications.

The research indicates that AI can enhance specialized knowledge, particularly for creative and highly educated workers. Still, it doesn't necessarily imply that organizations can hire individuals with minimal training in consulting or strategy work.

In the future, more studies will investigate which tasks are best suited for "Centaur" and "Cyborg" behaviors, emphasizing that humans should focus on tasks outside of AI's frontier of competence. The study suggests that finding the right balance between human and AI capabilities is essential for maximizing productivity and innovation.

As for the writing of reports like this one, it remains a matter of debate whether machines can outperform humans, indicating that the role of AI in various domains is still evolving.

4. SAP launches his own enterprise AI-agent named “Joule”

SAP has introduced its enterprise AI agent named Joule, which will be integrated into various SAP applications, including HR, finance, supply chain, procurement, and customer experience, as well as the SAP Business Technology Platform. Joule aims to enhance the SAP user experience by allowing employees to ask questions or frame problems in plain language and receive intelligent answers from SAP's business data and third-party sources while retaining context. For example, it can help manufacturers understand sales performance, identify issues in the supply chain, and suggest potential fixes.

Joule will be rolled out gradually, starting with SAP SuccessFactors solutions and the SAP Start site later this year, followed by SAP S/4HANA Cloud, public edition early next year. It will also be integrated with SAP Customer Experience and SAP Ariba solutions and SAP Business Technology Platform.

SAP's strategy includes building an enterprise AI ecosystem, with investments and partnerships with companies like Aleph Alpha, Anthropic, Cohere, Microsoft, Google Cloud, and IBM. Additionally, Sapphire Ventures LLC, backed by SAP, is dedicating over $1 billion to fund AI-powered enterprise technology startups.

5. ChatGPT has now access to the entire Internet

OpenAI's ChatGPT has expanded its capabilities to access the entire internet, providing current and authoritative information with direct links to sources through an integration with Microsoft's Bing search engine. This feature is available to ChatGPT Plus subscribers and ChatGPT Enterprise users, selectable through a drop-down menu.

This marks a return to web browsing for ChatGPT after a previous feature was disabled due to concerns about bypassing paywalls. The new browsing feature now recognizes the "robots.txt" code, which excludes certain content from web crawlers.

It's worth noting that Microsoft's Bing Chat, powered by a more powerful OpenAI language model than ChatGPT, also offers web browsing capabilities. However, using ChatGPT for web browsing allows users to stay within the ChatGPT interface with its additional features.

These updates follow OpenAI's recent announcements of ChatGPT's ability to analyze images, conduct audio conversations, and the introduction of the image generation model DALL-E 3.

6. Mistral AI release his first model for free and open source - and outperforms Llama

Paris-based startup Mistral AI has released its first large language AI model, Mistral 7B, which boasts 7.3 billion parameters and is open-sourced under the Apache 2.0 license. The model is designed for low-latency text summarization, classification, text completion, and code completion, and it outperforms Meta's Llama 2 7B and 13B models in various benchmarks, including language understanding, commonsense reasoning, and reading comprehension. While it falls short of outperforming the finetuned CodeLlama 7B in coding tasks, Mistral 7B offers memory and cost benefits for businesses, as it delivers the performance of a much larger model without affecting final outputs. The company plans to release a larger model capable of better reasoning and multilingual capabilities in 2024.

7. Hollywood writers’ strike ends with first-ever protections against AI

The Hollywood screenwriters' strike is ending with a ratified deal that includes protections against AI. The Writers Guild of America (WGA) reached a three-year contract extension with the Alliance of Motion Picture and Television Producers (AMPTP), allowing writers to return to work. The agreement includes increased pay, health and pension contributions, foreign streaming residuals, and viewership-based streaming bonuses. Notably, it establishes regulations for AI use, preventing AI from writing or rewriting literary material and ensuring AI-generated material doesn't undermine writers' credits or rights. Writers can use AI with consent, but it cannot be mandated, and AI-generated materials must be disclosed to writers.

8. The iPhone moment? John Ive and Sam Altman in talks

OpenAI, the company behind AI technologies like ChatGPT and Dall-E, is reportedly in talks with Sir Jony Ive, the designer of the first iPhone, as reported by the Financial Times. The objective is to create an "iPhone of artificial intelligence," a consumer device aimed at simplifying interaction and collaboration with AI in daily life.

This information comes from three insiders said to be familiar with the plans, and it gets even more intriguing. Masayoshi Son, the CEO of Japanese conglomerate Softbank, is also involved in these discussions. He intends to contribute approximately $1 billion to finance the project and advocates for using ARM chips in the device. This is a significant detail because Softbank owns a substantial 90% stake in the chip manufacturer ARM.

Jony Ive and OpenAI CEO Sam Altman have reportedly met multiple times at Ive's studio in San Francisco to discuss concrete ideas for this ambitious device. While a final design has not been established yet, some specific concepts are said to be on the table. In particular, the device aims to make interactions with AI more intuitive and shift the focus away from the traditional screen found on smartphones and similar devices.

Although the discussions are serious, the three parties involved have not officially signed a deal yet. According to the Financial Times, it might take several more months before an official announcement is made, and it could even be years before we see actual hardware.

While an "iPhone for AI" may still be a few years away, OpenAI could gain short-term value. This is due to an upcoming sale of shares by company employees who hold shares worth hundreds of millions of euros. Such a sale is likely to significantly boost the stock price. Currently, Microsoft holds 49% of OpenAI's shares, and other major investors like Sequoia Capital, Khosla Ventures, and Andreessen Horowitz also hold significant stakes in the company.

9. LIDL wants to compete with Amazon and Microsoft

Lidl's parent company, Schwarz-Gruppe, is entering the cloud business to compete with giants like Amazon and Microsoft. They've established a new digital unit called Schwarz Digits, aiming to attract companies with competitive pricing. Schwarz-Gruppe already opened its cloud to external businesses in early 2022, claiming to have hundreds of partner companies using their cloud solutions. With the core business facing challenges, they're expanding further into the cloud and IT sector, targeting competition with tech giants like Amazon and Microsoft. Schwarz Digits, with around 7,500 employees, will cover areas like cloud, cybersecurity, e-commerce, and retail media. The company plans to strengthen its position in IT solutions and digital services in the coming years, investing billions annually in strategic projects, including digitalization and IT. They emphasize the advantage of a European Union-based server location compared to Amazon and Microsoft and aim to attract customers with competitive pricing while fostering a "discounter spirit" in their new venture.

Note: The summaries for the Weekly Update was created with the help of AI.

🤝Digi-Pals other Socials: