As an AI language model, I cannot predict specific breakthroughs that will happen in 2023 since I do not have access to real-time data or the ability to foresee the future. However, I can mention some recent significant breakthroughs in AI and machine learning up until 2022. Please note that this list is not exhaustive, and the field is rapidly evolving:

1. Deep learning advancements: The continued progress of deep learning techniques has led to breakthroughs in areas such as speech recognition, image classification, and natural language processing.

2. Reinforcement learning achievements: Reinforcement learning has witnessed tremendous progress with milestone achievements such as AlphaGo and AlphaZero, which defeated world champions in complex games like Go and Chess.

3. Transformer architecture: The development of transformers, a type of neural network architecture, has revolutionized natural language processing tasks like machine translation and language generation.

4. Generative models: Advances in generative models, particularly through techniques like Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), have enabled realistic image synthesis, deepfakes creation, and data augmentation.

5. Transfer learning and pre-training: Techniques like transfer learning and pre-training models on large datasets (e.g., BERT, GPT) have improved the ability of models to generalize across tasks and domains, enabling transfer of knowledge from one domain to another.

6. Explainability and interpretability: Significant efforts have been made to enhance the explainability and interpretability of AI models, enabling better understanding and decision-making based on the inner workings of the models.

7. Edge computing and IoT advancements: The integration of AI and machine learning algorithms with edge devices and IoT (Internet of Things) systems has improved real-time decision-making, autonomous systems, and the ability to process data locally without heavy reliance on the cloud.

8. Autonomous vehicles: Continued progress in autonomous vehicle technology, combining vision, perception, and decision-making capabilities, has brought us closer to fully self-driving cars, improving road safety and transportation systems.

9. Medical diagnostics and drug discovery: AI and machine learning have played a valuable role in medical diagnostics, enabling early disease detection and helping to identify novel drug candidates, potentially speeding up the drug discovery process.

10. Ethical considerations: Increased awareness about the ethical implications of AI and machine learning has led to research and frameworks for ensuring fairness, accountability, and transparency, addressing biases and avoiding unintended consequences.

Please keep in mind that these are examples of breakthroughs that have occurred up until 2022, and the progress in AI and machine learning is ongoing.